Final Report Summary - OMSOP (Optimised Microturbine Solar Power system)
Executive Summary:
This project aims at the development and demonstration of concentrated solar power (CSP) parabolic dish system that generates electricity using a micro gas turbine (MGT). The technological objectives are to identify the components and overall system design requirements for producing an efficient reliable and cost effective system that can generate about 3 - 25 kW of electricity per unit using a single dish. The vision is to have modular systems that can be used either for distributed power generation or to be stacked in a farm arrangement to form medium size solar thermal power plants of several MWs of power, a range which is not economical if the commonly used steam turbine based CSP power plants are used. Parabolic dish CSP technology has been under development in recent years using Stirling engines as the power unit. However, questions have been raised about the reliability of these engines as well as their suitability for hybridisation with other fuels and thermal energy storage. Hence the unique feature of this project is the introduction of the micro gas turbine as the prime mover. Micro gas turbines have the potential to be lighter, more reliable, cost effective and flexible for hybridisation as well as the utilisation of exhaust heat in multiple generation maximising the overall efficiency of the system. The development of the demonstration plant incorporating a 6kWe micro gas turbine has been accompanied by the design and laboratory testing of short term thermal energy storage to smooth out fluctuations to variable solar irradiation. Additionally, techno-economic and worldwide market analyses have been conducted to identify the most suitable markets for the technology as well as the overall cost of the system upon volume production. Furthermore, system optimisation studies as well as life cycle analysis were performed to guide future developments of system.
The OMSoP project consortium is composed of eight partner organisations from five European countries from research organisations and SMEs with track record of research and project management in related areas. The project was organised into three technical work packages and a management work package led by City University of London with Professor Abdulnaser Sayma as the project coordinator. The technical work packages (WPs) are component development WP lead by KTH, system integration and demonstration WP led by ENEA and techno economic optimisation lead by RO3. The remaining project partners were INNOVA, responsible for concentrator design and installation, CP participating in the micro gas turbine control system and overall system optimisation work, USE working on market and cost analyses and European Turbine Network working on dissemination activities.
The project required 52 months to execute and resulted a successful demonstration of the solar powered micro gas turbine technology where the system was tested in Cassicia at the site of the Italian National Agency for New Technologies, Energy and Sustainable Economic Development (ENEA). Laboratory tests on the short-term energy storage demonstrated the capability of allowing smooth operation of the system. System optimisation studies produced several alternative arrangements for the power plants for future systems. The market and cost analysis showed that the OMSoP technology could be competitive with other solar power technologies in regions with high solar irradiation and favourable technological and economic conditions. The main conclusion from this project is that parabolic dish micro gas turbine technology is promising to be feasible, cost effective and reliable alternative to existing solar technologies and could open new markets. A framework should be put forward to develop the technology towards commercialisation.
Project Context and Objectives:
The main concept of generating electricity from solar insolation using a thermal power conversion unit is based on concentrating sun rays at a focal point where the temperature attainable rises to an adequate level for conversion to mechanical power. A working fluid, such as air, circulating through a prime mover that converts thermal power to mechanical power and then electricity, is heated at the focal point. Thus, a typical CSP system is made of a concentrator, a receiver-absorber and a prime mover. The concentrator is typically made of curved mirrors to concentrate solar radiation as much as possible and hence achieve high working temperatures. The receiver-absorber is a device located at the focal point which continuously extracts the thermal energy from the solar radiation. Part of the solar energy impinging on the receiver is transferred to one or more working fluids, while some of that energy is lost to the surroundings. Current small scale plants based on parabolic dish concentrators rely on Stirling engines as the prime mover or on concentrated PV. However, Stirling engines are known to have numerous technological problems affecting their reliability.
The utilisation of solar energy for power generation represents an important element of the renewable energy mix required to meet the environmental constrains in a suitable manner. Several technologies are currently being used to harvest the abundant solar energy such as photovoltaics and solar thermal power plants. Both technologies have made big strides in recent years in improving performance and in cost reduction. Parabolic trough collectors, or solar towers in conjunction with steam turbines are suitable, from practical and economic perspectives, for large scale power generation in the order of at least tens of Megawatts. Photovoltaics has been systematically reducing in price and are being deployed at scales from domestic applications to the Gigawatt scale plants, but back up power and energy storage remain major challenges for further deployment of such systems. A less developed solar thermal technology suitable for distributed power generation and power plants in the order of Megawatts is the parabolic dish plants that use a single dish to concentrate solar power at the focal point where the temperature of a compressed working fluid in a prime mover can be raised and the fluid is expanded to produce power. Both Stirling engines and micro gas turbines lend themselves as candidates as a prime mover for such systems. Parabolic dish systems using Stirling engines have been under research, development and demonstration for several years, however, their market deployment did not live to the original expectations due to several factors such as reliability and economy. Micro gas turbines as a prime mover in parabolic dish systems were trialled as early as the 1980s, but the unfavourable market conditions for renewables at the time were probably the main reasons that such technology was not taken further. Other recent attempts to integrate micro gas turbines with parabolic dish systems attempted to utilise of-the-shelf micro gas turbines not specifically designed for the purpose and were also much larger than could be sustained by a single dish leading to limited success. This did not lead to further developments. The Optimised Micro Turbine Solar Power (OMSoP) project was conceived to develop and demonstrate a purpose built micro gas turbine utilising a parabolic dish aiming to pave the way for successful market uptake due to realisation by the project team that specific features of such technology will make it suitable for applications and regions of the world more than other solar power technologies. This would be enabled by developing a reliable and efficient system with competitive cost of electricity in the target markets and application areas.
The vision of the OMSoP project has been to develop innovative concepts that can improve the ratio of the electric power generated to the solar energy insolation on the dish and to improve operability in relation to solar energy short time fluctuations through the development of short term thermal storage built within the receiver. These are important features that are expected to drive down installation and operation costs with respect to other solar technologies that should be accurately identified. Ultimately, the vision is that this should result in a solar dish-MGT system which is competitive to Photovoltaic (PV) and solar dish-Stirling engine systems. Subsequently, the aim has been to carry out a successful demonstration of a modular solar dish-MGT power plant which could in, a second phase, be deployed and taken up by the market in countries with high direct solar irradiation, such as those bordering the Mediterranean as well as the vast export markets such as China where the system is anticipated to be suitable to remote areas with no cost-effective provision of centralised power plants.
The minimum environmental impact of concentrated solar power (CSP) systems brings about the possibility to generate power close to the consumers, hence benefitting from eliminating transmission losses as opposed to other fossil fuel power stations. Due to the variation of demand depending on population density and the variable typical energy requirements per capita, a modular system which can be stacked to produce the local power requirements will be an ideal solution for rural and farming communities, but it can also be stacked in larger numbers to provide medium size power demand in the order of Megawatts.
The aim of the OMSoP project has been the development of plants that take current technology beyond the state-of the-art by integrating a highly efficient concentrator and receiver-absorber. The Stirling engine has been replaced with a micro gas turbine (MGT) with efficient power cycle which can be arranged with sub-cycles adequate for the power levels of the target application. The aim has been to develop plants that have performance and cost advantages that make them competitive with dish-Stirling and also with conventional fuel plants in certain markets; they also have the potential to compete with PV on dispatchability when integrated with thermal storage and/or co-firing using conventional or bio-fuels. Ultimately, this should pave the way for their widespread deployment in the future as a viable, reliable and cost effective clean energy source.
The ultimate goal of the OMSoP project has been to provide and demonstrate technical solutions which will allow future deployment of small scale reliable and cost effective concentrated solar power dish systems to enable industry to invest in the next generation of these devices. To achieve these objectives, it was required to address technological challenges with the primary components, the concentrator, the receiver-absorber and prime mover, leading to a demonstration of the concept.
In the concentrator, mirrors, made of plated glass or other reflective materials, typically contribute to the high initial cost of the solar dish systems, and, even if they have a high reliability, they can be too heavy, requiring robust positioning mechanisms and support structures. The primary challenge for the parabolic dish concentrator is the development of a new mirror technology that can ensure high performance at a competitive cost and an easily transportable size.
The main challenges for the receiver-absorber within the parabolic dish system were reliability and cost. Existing designs of the receiver make it one of the life-limiting components in the entire system. This is due to thermal stresses, thermal cycling, the use of brittle materials for absorption such as ceramic foams and the deteriorating efficiency of the sealing glass window under high thermal loads. Localised overheating and flow instabilities can also occur and these also serve to limit the effective life. Other challenges result from the high temperature that may result re-radiation of energy and significant losses to the surroundings. The aim here is to systematically address those challenges to meet the overall system performance, reliability and cost requirements.
A primary feature of this research is to replace Stirling engines, typically used in the target power range, with MGT. Gas turbines are known to have good weight to power ratio, reliability and ease of maintenance. However, downsizing existing MGTs to a size suitable for a power rating less than 10 kWe brings about several major challenges. The high rotational speeds required, typically in the order of 140,000 rpm, put significant pressure on the bearing design and rotordynamic stability, the latter is compounded by the need to the system to have wider operational range than conventional MGTs due to the variable solar irradiation. Other challenges are associated with the high turbine inlet temperature and reliable, light weight recuperator design, needed to achieve high MGT efficiency, and thus reducing the overall size of the concentrator. Such aspects have been addressed in this research.
Further studies aimed at the development of short term thermal storage technology that could be integrated with the receiver to smooth out short term solar irradiation fluctuations. Optimisation studies also aimed at identifying potential cycle arrangements and multiple generation concepts for maximum overall efficiency.
To enable commercial success, it has been also important to conduct detailed techno-economic study of the proposed system as well as to perform worldwide market analysis to identify the most suitable markets for future deployment. Both studies aimed to paving the way for future phases of the technology development toward full scale commercialisation. The target system has been expected to achieve the following outcomes:
- A high efficiency solar concentrator and receiver system that may include integrated short term storage capability and fossil fuel backup.
- Reliable micro-turbine system.
- Stand-alone machine.
- Modular, lighter and more compact system with simple cooling requirements.
- Low overall cost, easy installation and reduced maintenance.
- Potential for integration with medium and long term storage.
A life-Cycle Assessment will be performed including requirements for standardisation and integration with the European policy and legal framework, for economical assessment and market exploitation.
Project Results:
To achieve the OMSoP project objectives, a work plan has been designed which required the organisation of the project into three technical work packages (WPs) as follows:
Work Package 1: System component development, this included development of a solar concentrator, development of a solar receiver and development of a micro gas turbine. This work package including the following partners:
Kungliga Tekniska Högskolan (KTH) (WP leader)
City University of London (CITY)
INNOVA Solar Energy Srl (Innova)
Compower AB (CP)
University of Seville (USE)
Univertiy of Roma TRE (RO3)
Work package 2: System design and integration, this included overall system design, integration and testing representing the demonstration activity of the system in representative conditions. This work package included the following partners:
Italian National Agency for New Technologies, Energy and Sustainable Economic Development (ENEA) (WP leader)
CITY
RO3
CP
KTH
USE
Work package 3: Techno-Economic analysis, the work focused on two major aspects, system optimisation considering several alternative arrangements and market and cost analysis for future deployment. This work package included the following partners:
University of Roma TRE (RO3) (WP leader)
CITY
ENEA
CP
KTH
USE
The scientific and technological achievements of the project will be grouped below under the following categories: (1) the solar concentrator, (2) the solar receiver, (3) the micro gas turbine, (4) system integration and testing, (5) system optimisation and (6) market and cost analysis. More details can be found in the project deliverables available in the project web site (https://etn.global/research-innovation/projects/omsop/).
The solar concentrator
INNOVA conducted a study to identify the optimum arrangement for a concentraed solar power parabolic dish system arrangement. It was found out that a single dish with the power unit installed on the arm is the most appropriate for the target power range and for stacking several systems to form a power plant. The conducted studies on the dish structure and mirror materials suitable for the demonstration plant and finally performed a design of the dish and sun tracking system. The assembly of the dish was completed in April 2015 while the experimental characterization of the dish on-site started in May 2015. The dish characterisation was accomplished by ENEA trough two different methods: A) via simulation, mainly considering the experimental data of shape and solar reflectance of the facets composing the dish (indirect method); B) by direct experimental measurement.
The reflectance measurement of the Almirr mirror, which is the reflecting material selected by INNOVA, was performed at the ENEA labs.
The dish shape evaluation required two additional experimental steps: geometrical characterisation and facet canting. The geometrical characterisation of the dish structure was accomplished through the photogrammetry method.
The facet canting (mirrors orientation) procedure was carried out adopting a new experimental method developed by ENEA (Visual Inspection System method)
The facet canting took a couple of days; Figure 1.1 shows the progress from the beginning to the conclusion of the procedure. At the end of the canting a large part of the dish area appeared red-coloured (aligned); the remaining part is essentially black, meaning imperfection of the mirror shape.
Then the dish shape measurement was performed trough the VIS method to evaluate the slope deviation of the dish surface from the ideal one. At that purpose an innovative experimental set-up named VISdish was built up.
The outcome of the dish shape measurement is reported in Figure 1.2 where the slope deviation, peaked at 5-meter radius, is represented; from these results the limited optical quality of the dish was clearly noticed, most probably due to mirrors shape deformation during the on-site installation.
Based on the experimental data collected (mirrors solar reflectance and slope deviation), an accurate simulation of the concentrated flux around the focal point of the dish (Figure 1.3) was performed through the application of the SIMULDISH optical model, developed by ENEA. The dish optical efficiency detected was lower than the initial prediction and, accordingly, a resizing (enlargement) of the solar receiver was performed through the application of the developed system model to keep a constant power input to the MGT.
To validate the results obtained through the application of the SIMULDISH software (indirect method), the concentrated solar flux distribution was experimentally measured using an optical target positioned in correspondence of the focal point (direct method “Flux Mapper”, Figure 1.4).
The experimental tests were initially carried out during the night (moon tracking, Figure 1.5) to work in safe condition and get familiar with the instrumentation and subsequently were replicated on-sun, whose results are represented in Figure 1.6.
The outcome of the direct evaluation method was quite consistent with the indirect simulation, even if a worse dish performance was detected in the first case
A theoretical analysis was performed by ENEA to resize and propose new solutions for the realization of a new prototype of a solar parabolic dish collector to be integrated with the receiver/micro gas turbine (MGT) system developed within the OMSoP project.
Two new possible configurations were defined, starting from a critical analysis of the results obtained on the prototype installed at the ENEA Casaccia Centre. As a first option, it was assumed to have a system with the same electrical power output as the prototype (7 kWe), while in a second option a higher electrical power output (15 kWe) was assumed. Figure 1.7 shows a sample of the results.
The solar receiver
Two receiver concepts have been developed for risk mitigation: a robust cavity and a volumetric receiver offering a higher efficiency. For the demonstrator unit the more robust but less efficient cavity receiver has been chosen. Both receiver types were tested in the KTH high-flux simulator (HFSS) and based on the testing results the concepts were optimized for the use in the OMSoP solar dish demonstrator. This testing includes receiver component test as well as integrated testing of the receiver with a micro gas-turbine (MGT). Finally, the optimized cavity receiver was tested in the OMSoP demonstrator. RO3 has investigated the Short Term High Temperature Thermal Storage Receiver (STHTTSR) concept development.
For the demonstration plant a receiver concept that uses impinging jets to enhance the heat transfer was chosen. It was designed using an inverse design method as well as conjugate heat transfer analyses. Initial receiver component tests in the KTH solar simulator showed that the concept is promising for the integration in the OMSoP system. It was found that it is important that the absorber walls are fully oxidized before the receiver can be installed and operated in the real dish system. In a second step the cavity receiver was coupled to a MGT and the KTH solar simulator simultaneously to investigate the dynamic behaviour of the coupled system as well as to obtain improved measurement results of the cavity receiver performance. It was shown that the impinging receiver is suitable for the integration in the OMSoP solar dish system. During the experimental campaign air outlet temperatures of up to 810°C were obtained with absorber cavity material temperatures at around 960°C. The mechanical and life objectives were met with this design. It was shown that the total energy loss of the receiver is around 9.15 kW, and the efficiency of the receiver can be estimated to reach 73.7% at its design boundary conditions.
A receiver couples with short term thermal storage was developed and tested in the laboratories of RO3. The idea was to smooth out short term fluctuations of solar irradiation including the passing of clouds for periods of several minutes. The short-term Heat Storage Receiver (StHSR) concept shown in Figure 2.1 is based on a tube bundle system immersed in a Phase Changing Material (PCM). The container has a front plate that is directly irradiated by the concentrated solar energy. The heat is transferred into the PCM by means of the back of the frontal plate and solid heat pipes (acting as fins), thermally connected with the frontal plate and immersed in the PCM that provides for the heat transfer to the working fluid. During the heat storage, the liquid phase fraction increases because the entering heat power is higher than the exit one. When, due to the DNI fluctuations and partial or total solar eclipses, the entering heat power becomes lower than the exit one, because of the latent heat released by the PCM solid fraction increase, the working fluid exit temperature variations are damped in respect to that caused by direct irradiation of the tube bundle.
The preliminary design has led to select a tube bundle system made of 12 tube bundles to be connected by an external muffler as shown in Figure 2.2. This solution has been chosen to use the same receiver for complex multi-shaft MGTs with multiple heating to achieve high efficiencies of the same order of the Stirling engines for CSP applications.
Since the level of knowledge on such kind of receiver is almost non-existent, a theoretical model based on a FV approach with lumped phenomena in the volume nodes has been developed to allow for implementation of the not yet known phenomena that would be evidenced by the experimental activity.
The short-term heat storage receiver has required tests on models to understand how the melting and the solidification occur inside the volume and to understand the heat transfer process. The models have been tested in the RO3 solar lab using a spiral tube bundle (Figure 2.3) where xenon lamps, reflectors and Fresnel lens have been used to simulate the concentrated sunrays. Figure 2.4 shows a view of the test facility. Tests were carried out by filling the container by a NaNO3/KNO3 60/40 salt provided by ENEA with the front plate having a Smooth back Surface (case SS) and equipped with solid heat pipes (SHP). Results showed that the SHP increase strongly the heat transfer rate. The frontal surface thermal temperature has been measured by a pyrometer. Further tests have been carried out by measurements made by thermocouples immersed in the charge. The front plate temperature distribution has been measured also by a thermos-camera. The hot spot and the temperature distribution along one plate diameter are shown in Figure 2.6. The temperature measurements versus time are shown in Figure 2.7. These experiments allowed the calculation of the power impinging entering the frontal surface of the plate and the emissivity of the same surface. Moreover, the heat power transferred from one volume to the other has been evaluated.
A Finite Volume Model based on lumped parameter concept, sketched in Figure 2.8 has been developed for simulation purposes. The model is flexible because various phenomena can be activated according to the experimental evidence. Simulations reported in Figure 2.9 have evidenced similarity in the front plate tested temperature shown in Figure 2.6 and great capability of the receiver to soften the oscillations induced by a DNI step change and sinusoidal oscillations. An example is given in Figure 2.10.
Different tube bundles have been evaluated. Such tube bundles were arranged as spirals, U tubes and Helix Bundles. The last ones have demonstrated the best behaviour and then the final arrangement has been sized accordingly. Twelve tube bundles are required to cover, by parallel and series multiple connections, various MGT arrangements that, in the perspective, they can be used to satisfy market requirements. LiF has been selected as PCM for 800°C MGT TIT and MgSi has been selected for 900°C TIT. The weight of the LiF PCM in the receiver to supply heat power for 45 minute with zero DNI is about 43kg. Longer time interval of full heat power supply by the receiver increases the receiver weight that requires it would be installed on the ground.
The performance of the cavity receiver was evaluated in the KTH HFSS. Figure 2.11 shows the front (A) and side (B) view of the solar receiver installed in the HFSS.
To calculate the receiver performance indicators pressure and temperature measurement have been taken at different positions. In Figure 2.11 can be seen pressure transducers (PT), K-type thermocouples (KTC), and an IR-camera. Compressed air at approximately 20°C enters the receiver at the side through a measurement pipe which is fitted with thermocouples and pressure transducers. The mass flow can be adjusted remotely using a mass flow controller upstream. Inside the receiver, the air is heated by concentrated radiation supplied by the solar simulator before expanding in the choked nozzle which connects to the ventilation system using a flexible high temperature hose.
Initial tests of the cavity receiver were performed without integration of a MGT and without the electric air heater which means that the air entering the receiver was approximately 20°C. First, the non-oxidized receiver was gradually warmed by 4 lamps, 8 lamps and finally 12 lamps with an air mass flow input of 30g/s at an absolute pressure of 3 bar. After one hour of testing under the full power of 12 lamps, it was observed that the bottom region of the receiver was significantly oxidized which happens at a temperature level above 800 °C, as shown in Figure 2.12 even though the thermocouple on the cavity wall showed that the temperature on the cylindrical surface only reached approximately 340°C. It was concluded that more radiative energy is absorbed in the bottom region of the cavity as compared to what it was designed for.
As the cavity receiver was designed to operate at temperatures of 800°C and higher, the cavity material temperature reaches values above 800°C which means that the material (especially metallic material) surface would be fully oxidized. During the design process, the emissivity of the material with a fully oxidized surface was used instead of the emissivity of the material with a polished surface. For a receiver with a non-oxidized absorber surface the radiative boundary condition could be totally different to the design conditions. Normally, an oxidized surface has a higher emissivity than the metal itself. Therefore, the high flux region on a non-oxidized absorber is located further towards the aperture than on a same sized oxidized absorber due to the multi-reflection inside the cavity. The bottom of the non-oxidized absorber, where no impinging jets are distributed, might work under an undesirable high light flux level and overheat condition. Moreover, the cylindrical wall region located under the impinging cooling jets works at a lower temperature than the design temperature and might work below the oxidation temperature. This phenomenon might cause receiver failure if it would be fully focused without pre-oxidation directly after installation on a real dish system. The results imply that for using the impinging solar cavity receiver the real dish system it is imperative that the absorber is fully oxidized before installation.
Before the following detailed testing, a pre-oxidation process of the absorber was attempted by using an Acetylene burner. After the oxidation process, the cavity absorber is closer to a black body than the cavity before testing and the cavity after one hour testing with 12 lamps, as shown in Figure 2.12 (C). However, a full oxidation could still not be accomplished on the side wall surface which means that the optical efficiency is reduced due to higher reflectivity on the side walls.
The pressure drop across the solar receiver remains well below 0.1% for the operating points investigated. It can be seen that the total efficiency decreases significantly from 53.0 % to 42.9 % with the decrease of the mass flow from 30 g/s to 15 g/s. Due to the decrease of the mass flow, the heat convection efficiency on the back-side of the receiver also decreases. This in turn leads to the increase in wall temperature as seen in the experimental measurements of the wall temperature (the larger value measured by the two thermocouples on the wall). It increases from 329.6 °C to 435.2 °C with the decrease of the mass flow from 30 g/s to 15 g/s. Therefore, both the radiation and convection heat losses increase and finally cause the total efficiency to decrease. At nominal mass flow operation and design radiative boundary conditions the efficiency is calculated to be 73.7%.
Analysing the test, it can be concluded that the cavity receiver concept using impingement jets is promising for the integration in the OMSoP system. It is important though that the absorber walls are fully oxidized before the receiver can be installed and operated in the real dish system.
In parallel to the design and testing of the cavity receiver a volumetric receiver concept was designed and a prototype built for experimental evaluation in the KTH HFSS. To obtain representative results when testing the volumetric receiver in the solar simulator scaling effects were investigated between a receiver installed in the OMSoP solar dish and the KTH HFSS. This study is based on a coupled CFD/FEM analysis in combination with ray-tracing routines that have been validated against measurement data of the OMSoP solar dish and the solar simulator. Because of the analysis, a scaled prototype was manufactured.
The receiver was placed in the focal spot of the HFSS and connected to a high-temperature air heater at the inlet and a chocked nozzle to simulate the MGT at the exit as shown in Figure 2.13. Compressed air is supplied by a high-pressure compressor and the mass flow is adjusted by a mass flow controller before being heated by the electrical air heater to 540°C. Concentrated radiation provided by the HFSS heats the air inside the volumetric receiver before being expanded in the choked nozzle. Before conducting the experiments the receiver was covered with approximately 15 cm of high temperature insulation.
To calculate the receiver performance indicators pressure and temperature measurement have been taken at different positions. In Figure 2.13 can be seen connections for the pressure transducers (blue tubes), K-type thermocouples (green connectors), 4 pyrometers mounted in a 2-axis linear unit in front of a water cooled shield. The pyrometers are used to measure the receiver glass window temperature as well as the volumetric absorber temperature during operation. For the measurement, the sensors which are protected by the shield are moved quickly in front of the receiver with the lamps still operating while the air mass flow is turned off.
Tests have been performed reaching a maximum air outlet temperature of up to 820°C. At the nominal operating point the air temperature inside the receiver was increased from 540°C to 800°C with a conversion efficiency of approximately 80%.
Analysing the test it can be concluded that the volumetric receiver concept is a suitable candidate for the integration in the OMSoP system with a measured efficiency of 80% and the potential to reach even higher values.
The cavity receiver was coupled to a MGT supplied by CP in HFSS KTH lab in order to investigate the dynamic behaviour of the coupled system as well as to obtain improved measurement results of the cavity receiver performance.
The MGT provided by CP was coupled to the receiver mounted in the KTH HFSS. The MGT has a nominal electrical power output of 5kWe, a nominal speed of 160,000 rpm, requires a thermal input of 31.25 kW, and operates at a compressor pressure of 3 bar (absolute), with a nominal mass flow of 100 g /s with a turbine inlet temperature of 800°C.
Figure 2.14 shows the layout of the integrated MGT and receiver. Air is compressed in the compressor to 3 bar (absolute) before entering the heat exchanger (HEX) where it is heated to 500°C. Then the concentrated radiation heats the air within the solar receiver to 800°C or higher before being expanded in the turbine. In a next step, the air passes through the combustion chamber where additional heat is supplied to the system if necessary. Finally, the air passes through the heat exchanger and is then expelled at the exhaust.
The physical integration of the receiver and the MGT required a couple of meters of piping as shown Figure 2.15 Figure 2.16 and Figure 2.17.
Since the prototype receiver is originally designed for different radiative boundary conditions, the HFSS can only provide 30% - 42% of the radiative power required. Therefore, two methods have been applied to achieve the required absorber temperature during the testing: increasing the receiver inlet temperature and reducing the mass flow.
During the experimental campaign, multiple cases with different start boundary conditions were investigated and six representative ones are presented here. To avoid the quenching effect caused by the impinging jets, the MGT and the HFSS were turned off at the same time during the cooling down operations. Figure 2.18 shows the experimental results in terms of the absorber temperatures, receiver inlet temperature, outlet temperature, air mass flow and radiative power from the HFSS.
During most of the time the MGT ran at a low speed and the mass flow was controlled between 22 g/s and 26 g/s to achieve the required absorber surface temperatures and receiver outlet temperature. It was found that the receiver can produce outlet air of up to 810.5°C with absorber peak temperatures in the region of 960°C which match closely with simulations of the to the receiver performance in the OMSoP solar dish. Therefore, it can be assumed that the energy losses and cooling down rates are representative when compared to the solar dish system. During start-up it was found that temperature differences in the absorber cavity reached values of up to 100°C when the receiver is heated from ambient temperature. To avoid structural problems, it was concluded in the future real applications, this ’cold start’ should be avoided.
Additionally, to avoid thermal shock and to guarantee lifetime the temperature changing rate of the absorber cavity is important. It can be seen in Figure 2.19 that the absorber temperature and the receiver outlet temperature change quickly during the start-up and turn off.
In cases 2, 4, 5 and 6 the absorber temperatures have reached quasi steady states at similar levels compared to the receiver design temperature condition. Therefore, the energy losses of the receiver can be used for predicting the receiver total efficiency. Figure 2.20 shows the details of the energy losses for these four cases. It can be seen that the heat losses are significantly affected by the mass flow (see the curves of case 4, 5 and 6) as higher mass flows help to take up more heat by offering a higher velocity and Nusselt number. Material temperatures are another factor that affect the energy losses due to increased radiation and convection heat loss (see the curves of cases 2 and 6).
Considering the data stability, the average energy loss value of case 2 is used for predicting the receiver efficiency when it works under its designed boundary conditions. With the measured heat losses, the efficiency of the cavity receiver can be estimated to be around 73.3% for a solar that provides 35.3 kW in an aperture of 200 mm.
Analysing the integrated MGT-receiver test it can be concluded that the impinging receiver is suitable for the integration in the OMSoP solar dish system. It was shown that the impinging receiver can reach air outlet temperatures of 810°C at an absorber temperature of 960°C. The absorber cavity temperature uniformity is good with exception of the ‘cold start-up’ process. The total energy loss of the receiver is around 9.15 kW, and the efficiency of the receiver can be estimated to reach 73.7% at its design boundary conditions.
The micro gas turbine
At the start of the project, it was postulated that an existing micro-gas turbine (MGT) developed by CP will be adapted for the CSP system. This 7 kWe MGT was originally developed for external firing for combined heat and power applications allowing for fuel flexibility. The turbomachinery components are based on modified turbocharger components and the shaft bearing arrangement are of the cantilevered type where the turbomachinery is over-hung from one end with the high-speed generator placed between two bearings.
Initial studies on the system operation and rotor dynamics showed that this system will have a critical vibration mode close to the operating range. In a typical operation, the MGT will be accelerated fast past that mode and operated above the critical speed. With the variable solar irradiation, it was deemed that it is possible that the system will be frequently passing past that mode during operation which may lead to high vibrations and failure. Alternative arrangements and designs were considered and it was decided to design a micro gas turbine operating sub-critical, will all vibrations mode above the operating range. The final arrangement is shown in Figure 3.1 with the generator placed in between the turbine and compressor. A new turbine and compressor design were completed and turbine materials were selected to comply with the subcritical operation. All the necessary electrical and electronic equipment were designed and built including the control system. The main S & T achievements are:
A new turbine design was completed which was optimised for high efficiency and light weigh. This we achieved using an in-house design model which followed by design improvements informed by computational fluid dynamic, structural dynamics and thermal analyses. These aim at meeting the design requirements while ensuring the elimination of host spots and high stress concentration points. The turbine material chosen was titanium aluminide (TiAL). Figure 3.2 shows views and dimensions of the final turbine design while figure 3.3 shows pictures of two manufactured turbines, the left-hand side one is an Inconel turbine manufactured using an additive manufacturing method while the right-hand side is machined from TiAL. Tests on the turbine were conducted at the City University micro gas turbine rig built during the project to verify the predicted behaviour. Figure 3.4 shows a comparison of experimental data and predictions with good agreement.
A set of electrical and electronic equipment were designed and sourced/built that were necessary for laboratory testing as well final integration with the solar dish system. These include a rectifier, an electrical drive and a sign wave filter that would allow motoring mode to higher speeds than in conventional drives to provide enough air protecting the receiver during start up model. The turbine test rig and electrical equipment are shown in figures 3.5 and 3.6.
Design and testing of the receiver interface and recuperator including pipe network and heat exchanger was completed as shown in Figure 3.7. CFD tools were used to provide guidance to the design process. The results were used by ENEA to integrate the system for the demonstration phase. Flexible pipes were used to connect different components, i.e. compressor, turbine, receiver and recuperator. This is mainly to accommodate the thermal expansion in the structure.
A novel solar operated micro gas turbine control system was developed, programmed and used for the demonstration activity. The preliminary control strategy, limiters and control sequences have been identified by CP. A dynamic simulation of the system was performed which showed the effectiveness of the proposed closed-loop control strategy. The proposed control logic facilitates the electrical conversion system to control the operation of the MGT. Further work was done by City, ENEA and CP to enhance the control sequence and the electrical scheme of the system which is the main mechanism to control the system. The hardware of the controller was programmed at City and tested later at ENEA in the demonstration plant.
The Micro Gas Turbine has been designed by City and manufactured partially at City workshop. The manufacturing of most of the components have been outsourced to different firms. The assembled City-MGT was prepared for the demonstration test of at EAEA site in Casaccia Centre.
A thermodynamic model of the MGT-CSP cycle was developed by City team and integrated in an optimisation framework to optimise the thermos-economic of the systems. An innovative operation strategy was developed and studied to improve the performance of solar only MGT-CSP. Figure 3.8 shows an example of the optimisation results for a MGT-CSP system to maximise the annual generated electricity while maintaining the cost of produced electricity on its lower level.
Comprehensive rotordynamic study was carried out by the City team to optimise the dynamic stability of the MGT. three different mechanical arrangements were compared for the MGT design. The study included using of alternative material for the turbine wheel to improve the dynamic stability. Table 3.1 summarises the pros and cons of different mechanical arrangements adopted for the MGT.
The turbomachinery components were designed and optimised for solar applications where due to the variation in DNI the off-design performance of the compressor and turbine become more critical. The CFD analysis of the compressor was integrated with an optimiser to enhance the performance of the system over a wide range of solar irradiation. Figure 3.9 shows the positive effect of compressor optimisation for the part-load operation on the overall cycle efficiency.
A detailed flow study was carried out at City to understand the effect of fine-scale features on a radial turbine on improvement of its aerodynamic efficiency. A test rig was setup and a series of experiments were carried out which showed the positive effect of the proposed features in the efficiency of small radial turbines. Figure 3.10 shows the effectiveness of the proposed features on the radial turbine efficiency.
System integration and testing
Based on the outcome of the experimental characterization of the components, the limited optical quality of the dish, and the results of the system simulation, the design of the demo system was iteratively revised, both in terms of component specifications and integrated system configuration. Regarding the component specifications, the receiver was resized and the system parameters were revised to keep unchanged the target power of about 5 kWe even in presence of an unexpected low optical performance of the dish. About the system configuration, different aspects were considered, ranging from mechanical to electric and electronic integration. For the mechanical aspect, the focus was on the design of a supporting structure which could meet the specification of every single component, ensure the mechanical stability of the system, and guarantee a free thermal expansion in every different inclination angle (Fig. 4.1).
Furthermore, a big effort was spent by ENEA, with the contribution of City and CP, on the design of the integrated electrical schemes and equipment. This aspect was initially underestimated and caused a delay on the completion and realization of the demo plant. Several electrical schemes were conceived mainly to individuate the best way to connect the electrical output of the MGT and to design the Power Electronic device in charge of speeding-up the Micro Gas Turbine and to transform its electric output into a usable electric flow. This work played a significant role in the project development since the different choices impacted on the controller hardware design and realization. The final solution involved the adoption of 4 individual components, connected according to Fig. 4.2 to realise a new conversion system, not commercially available.
The main system equipment (receiver, MGT, diffuser, and recuperator) were delivered to ENEA in winter 2017 while auxiliaries, oil pump, tubes and connectors were available in the first months of 2017. The process of physical assembly in lab took about four months, ending in April 2017, since several adjustments and modifications were conceived on-going leading to a system mechanical assembly more functional than the initial design. In particular the supporting structure was completely revised once started the process of assembly, to guarantee at the same time the mechanical stability of the system and the free thermal expansion of the hot components (mainly the receiver). Furthermore, all the connections between the components were studied to ensure the assembly and disassembly operation in case of damage of some components. The supporting structure was realized with 4 mm thick stainless steel L profiles and each component was anchored taking into account different constraints.
Initially the MGT was meant to be sustained directly by the recuperator but during the assembly process it became evident that the LP inlet flange of the recuperator could not bear the MGT weight, since the casing of the recuperator was significantly deformed once connected the MGT. Therefore, a dedicated system was developed to hold the MGT/HSG group to the structure.
The final system configuration is represented in Fig. 4.3 where the insulation of the receiver and the recuperator are also documented.
The entire system was installed inside a metallic container, thermally divided in two sections, the hot and cold areas to protect the compressor, the HSG, the oil pump and the electrical equipment from the heat absorbed by the receiver.
The installation of the equipment on the dish arm was performed in June, to allow the instrumentation, calibration and preliminary testing of the whole system in lab. Initially the receiver/MGT group was lifted and anchored to the dish arm (Fig.4.5 a). Then protective panels around the receiver/MGT group were installed to create a protective envelope from environmental agents (Fig. 4.5 b).
After the installation of the MGT group, the experimental procedure for aligning the receiver window with the dish focus was replicated by ENEA since the weight of the System Container altered the previous centring.
From the electrical/electronic point of view, the integration of the receiver/MGT group with the dish required the connection of the DCU (Dish Control Unit) with the MGT controller (LabJack unit) and the installation of four electrical boxes outdoor, housing the system power supply, the Drive, the Rectifier, the Sinus Filter, the Data Acquisition System and the Controller hardware. The cabling of all these units was a little troublesome due to the different components into which the Converter Unit is subdivided. This fact led to long connections of the power and signal cables, with additional pathways for many types of electrical noise either from e.m. radiation and ground current paths.
For the demonstrative plant, which is aimed at verifying the technical feasibility of the integration between the solar dish and the MGTs, a sophisticated Data Acquisition System (DAS) was realized by ENEA, in collaboration with City and KTH), to deeply understand and evaluate the performance of this innovative technology. The DAS conceptual scheme is represented in Fig. 4.7 where the presence of two separate systems (DAS1 and DAS2) is put in evidence: DAS1 is aimed at acquiring the signals from the environmental sensors, while DAS2 is destined to register the process parameters. Furthermore, a third parallel system of acquisition was developed to directly send the value of the most important and critical process parameters to the controller (Labjack unit). The installation of the instrumentation took three months, starting from April to June. More than 50 process sensors were purchased by ENEA and City and installed on the demo plant, partly on the receiver/MGT group, partly on the electrical equipment, as represented on the scheme in Fig. 4.8 where the parameters acquired for control purpose are highlighted in red, while the signals destined to registration and experimental analysis are highlighted in green.
The calibration of the instrumentation was carried out contemporaneously with the sensors installation. Each sensor measurement chain was checked to verify the transmission of the right physical variable value. Even if the calibration procedure was completed in lab before June, once the System Container was installed on the dish and the demo plant operated, a further revision was necessary due to the presence of high and low frequencies noises on the signals, mainly generated by the inverter of the dish and the Drive. In particular the voltage signals coming from the Controller Remote Unit (Labjack) were significantly affected by the noise. Corrective actions on the grounding of the equipment were taken and RC filters were successfully installed on the Labjack module.
The testing on the demo plant at the ENEA Casaccia site started in the early 2016 to verify the resistance of the solar shield, a panel to be positioned in front of the OMSoP System Container to protect mechanical and electrical devices from the solar reflected radiation (Fig. 4.9). That component was initially foreseen but became a critical issue when, from the experimental flux distribution characterization conducted in the years 2015-1016, it was evident that the high concentrated solar radiation reflected by the dish only partially was intercepted by the receiver window, and a large part was spread around the focus.
In a second step, after the completion of the system assembly and instrumentation, a first series of tests was performed in lab to verify the correct operation of the sensors installed and to check the ability of the Drive to control the MGT speed rate. The speed rate of the MGT was increased progressively during the tests, from 40000 rpm (Fig. 4.10) up to 75000 rpm (Fig. 4.11) to stress the system gradually, to acquire awareness of each single aspect related to the system functioning. Some oil leakages were detected during the operation of the MGT, particularly at low speed levels. Corrective actions were taken by City and ENEA to fix the problem. Anyway, based on the experience gained on the management of the oil circuit, it was evident that an oil-free design of the MGT must be developed as a future improvement of the technology. From the experimental campaign it was also confirmed that vibrations get lower (Figure 4.11) when increasing the speed rate over 70 krpm: therefore it was concluded that during the start-up procedure of the demo plant the speed rate has to be brought very quickly over the 70 krpm threshold to reduce the mechanical stress on the MGT. Many other information and data were collected to guide the operation of the system when tracking the sun.
Due to the late installation of the MGT/receiver group on the dish arm, mainly caused by the delayed definition of the electrical schemes and the delayed delivery of the electrical equipment (June), control hardware and software (beginning of July), the on-sun testing activity was limited to the month of July. ENEA, City and CP collaborated in the definition of the operational sequence, both in automatic and manual modes. Initially the system was operated in off-sun position, replicating the same tests as the ones performed in lab. The focus was put on the evident signal interference occurring during the dish manual operation on the parameters acquired by the Labjack module: the range of variability was relevant, even in cold test conditions, and some parameters presented evident off-sets values. As a corrective action, a module of 14 RC filters was connected to the signals entering the Labjack unit to protect them from the high frequency noise coming from the inverter of the dish, the inverter of the air fan and the Drive.
As a first test, the dish was put on automatic tracking and the MGT was motored up to 70 krpm to verify the ability of the control software to command the single components, included the dish. Nevertheless, an off-set of 20° on the dish azimuth position was set to prevent the heating of the system and perform a test out of focus. The MGT speed rate analogue signal acquired by the control unit (Labjack) still presented a significant variation from the set value, due to an improper grounding of the Drive, which was subsequently rectified. From the subsequent “cold” tests it was confirmed that the critical MGT speed range for the vibrations is 50-60 krpm (Figure 4.12).
Once verified the full operability of the different system apparatus mechanical/electrical/control), the experimental campaign on-sun was started with the contribution of City.
The testing procedure was structured in two main steps:
an initial tracking phase with an off-set of 20° in the dish azimuth to allow the system to reach high speed rate in cold conditions and to check the functionality of each equipment
a full tracking phase, with the receiver window aligned with the solar spot.
In the first tests, in a cautionary approach, the receiver was put intermittently on focus to allow a gradual heating of the receiver’s walls and absorber. Some cloudy days helped in limiting the heating ramp of the receiver (Fig. 4.14) where the evident difference between the TIT and the receiver outlet temperature (RECV-T2) can be attributed to the thermal inertia of the flanges and the piping connecting the receiver and the turbine.
In Figure 4.15a and Figure 4.15b the temperature courses and the pressure drops detected along the recuperator are represented, respectively. Obviously, the pressure drops tend to get higher while increasing the TIT.
The pressure levels detected along the circuit are represented in Figure 4.16: the compressor outlet pressure is within the range 1.00-1.47 bar in the MGT speed range 0-80 krpm. The vibration levels were quite stable and well below the limit of 7 mm/s; only isolated peaks were detected in correspondence of abrupt variations of the MGT speed rate. Therefore, as a recommendation, the MGT speed rate has to be gradually modulated during the start-up operation.
Each test was stopped once the turbine bearing temperature approached the value of 100°C. In this regard, the turbine bearing valve in the oil distributor was fully opened to increase the oil flow rate and consequently the cooling effect on the turbine bearing. The incremented oil flow led to a reduction of the turbine bearing temperature with respect to the compressor bearing in cold conditions (up to 90 krpm). Differently from the tests performed in lab, the stator temperature did not increase sharply, most probably thanks to the adoption of a finned heat exchanger in the lubrication oil circuit. Other on-sun tests were performed with the aim of registering the system parameters (both process and electrical parameters) in different environmental conditions and identify possible criticalities to be overcome (Fig. 4.17 Fig. 4.18).
Some unexpected difficulties were encountered, such as a relevant oil leakage occurring during an on-sun test: From an in-depth examination of the oil circuit it was verified that the cap gasket of the oil tank was deteriorated, leading to copious oil leakages when the dish inclination was higher than 45° from the horizontal axis. The gasket was replaced with a new one and the problem was fixed.
During the experimental campaign, it was noticed that mechanical power was produced even at low TIT levels (100-300 °C): in correspondence of a MGT speed rate of 65-70 krpm and a TIT range of 100-250°C, the power produced was about 500 W. (see Fig. 4.19). In correspondence of 80 krpm the maximum power produced in the TIT range 100-270°C was 900 W (Figure 4.19). At lower speed rate levels the production of mechanical power was less evident, even if not negligible.
Based on a linear extrapolation of the 80 krpm curve, the mechanical power that can be produced by the Turbine of the MGT in correspondence of 800 °C is expected to be 4.5 kW. Unfortunately, it was not possible to experimentally verify the validity of such prediction and to characterize the system behaviour at TIT levels higher than 300°C (and receiver temperatures higher than 700°C), due to the unexpected overheating of the turbine bearing temperature in correspondence of TIT higher than 300°C. Indeed, from the experimental campaign on the demo plant it clearly emerged that the turbine bearing temperature increases very steeply once the system starts to heat up, following the TIT course. As an example, in figure 4.20 the temperature course of the turbine bearing (BER_T2) detected during an on-sun test is represented, along with the TIT behaviour. The heating ramp of the TIT was quite steep in this test, since the associated DNI was above 800 W/m2 (Fig. 4.21). After some minutes the dish was put out of focus to avoid the turbine bearing overheating. The bearings adopted in the present MGT design cannot stand temperatures higher than 120°C. Some corrective measures were undertaken to efficiently remove the heat from the bearings, such as increasing the oil lubrication flowrate and pressure, but no evident results were obtained. Indeed, even if the oil lubrication inlet temperature was kept substantially constant during the tests, thanks to the adoption of a finned heat exchanger, the cooling effect in the turbine bearings was not sufficient. Due to time constraints, it was not possible to rectify this issue within the project duration to increase TIT levels.
However, the operability of an innovative system which can generate electrical power from solar irradiation was proven through the experimental campaign (Fig. 4.22). The management of the operating procedures through the control software was successful and the mechanical and thermal resistance of the components and the whole assembly was verified up to a speed of 90 krpm, a TIT of 300 °C and receiver temperatures up to 700 °C. Empirical system parameters, such as critical speed rates (50-60 krpm), duration of the defocusing procedure (15 seconds), thermal inertia of the components, lubrication pressure (2,5-3 bar), were identified. Based on the results of the experimental campaign, an in-depth analysis of the system performance, both in terms of stationary and transient behaviour, will be performed in the future. Furthermore, with the knowledge gained through the operation of the demo plant, several recommendations in view of a future improved design were identified.
System optimisation
The steady state plant simulator for design purposes has been developed using the models and approaches presented in the previous periods. The plant simulator contains the various steady state modules associated to components as shown in the Figure 5.1.
The Model of the environment establishes the air quality (dry air composition and relative humidity), temperature and pressure. Moreover, the sun DNI is defined.
It is considered as a parabolic shaped solid dish as shown in Figure 5.2 it is characterized by sizes: diameter, reflecting surface, by optical performance: efficiency, maximal temperature of the spot related to the DNI value and shape of the spot. For this application, a perfect tracking system is considered, thus, the tracking error is set to zero.
The receiver is the device that transfers the radiant heat power concentrated at its opening to the Working Fluid entering it, and to the environment as lost heat power. The component model is sketched together with an example of cavity receiver in Figure 5.3.
The MGT module contains the MG selection and sizing module that had already been completed. It is based on the block diagram of Figure 5.4 that allows to establish the sizes of turbomachinery and the related maps according to the impeller technology database, an example of such a DB is shown in Figure 5.4. The engine related equation set represents a correlation among performance and sizes of subcomponents. The internal quantities are also calculated to be used for other purposes. Various complex MT arrangement can be selected.
Figure 5.5 shows the relationship between efficiency and electric current for the whole electric subsystem. The section concerning the auxiliaries leads to establish the power absorbed by the auxiliaries on empirical basis.
All these module equation sets are solved by a modular approach. The degrees of freedom are optimized according to the minimization of a suitable objective function. The scheme of the process is shown in Figure 5.6. It allows the cycle calculation using the calculated efficiencies of the components by an iterative procedure. After the components have been sized, their off-design behaviour is established. Using all the off-design modules the matching problem among them is solved. Results represent a steady state running point of the plant, the boundary conditions being assumed.
For the OMSoP demonstration plant, the Figure 5.7 shows the part load behaviour versus the DNI. The left-hand figure gives the produced power, the CSP efficiency, the MGT rotational speed and the mass flow rate. The right-hand figure gives the electric efficiency, the receiver efficiency and the product of the MGT compressor, expander and mechanical efficiencies.
A comparison between the Stirling CSP performance and the MGT-CSP ones has been done. Stirling CSP efficiency is given versus electric power in Figure 5.8. The above steady state simulator has allowed the evaluation of the various CSPs equipped with the complex MGT arrangements. Results are shown in Figure 5.9 it shows that for both Stirling and MGT, the higher the power is, higher the efficiency is. Increasing the MGT arrangement complexity, especially for the higher power (35kW), the MGT-CSPs have better expected performance than the Stirling CSPs.
To improve the overall CSP plan performance reducing costs, some MT-CSP combinations with ORC have been analysed.
First, a complete investigation focused on ORC cycle net efficiencies has been carried out checking in the real applications. Figure 5.10 shows the ORC net efficiency versus the temperature of the heat source. Various combinations of the ORCs and the CSP-MGTs have been studied taking as reference the OMSoP demonstrator.
A first combination is shown in Figure 5.11 that shows an ORC cycle bottomed to the MGT. Since the MGT exhaust can have a temperature that ranges from 250°C to 300°C, an average ORC efficiency of 13% can be expected. Thus, the ORC turbine can produce some 3kW. The demonstrator global power could become about 9kW and the overall efficiency can increase of about 44%.
A second proposed combination scheme is depicted in Figure 5.12. It is for a combination between the MGT and an ORC that takes the heat power from the MGT exhaust and from the focal spot by a dedicated tubular receiver being irradiated by the part of the spot, whose irradiance levels lead to temperatures lower than that required by the MGT receiver. This heat power can be stored at medium temperature suitable for the achievement of good ORC efficiencies. Such a stored energy can be used to compensate the sun irradiance fluctuations that influence the TIT and the MGT power output.
Another proposed arrangement has been analysed for 24h working by mean of a medium temperature concentrator and thus lower cost per kW of heat power has been foreseen. The combination of the two cycles can lead to operate the plant 24h per day with an average efficiency intermediate between the combined MGT – OR simply combined cycle and the ORC one.
Market and Cost analysis
The economic performance of the OMSoP system can be assessed through several variables, the most relevant of which are the Installed Cost of the system (€/kWe) and the Levelised Cost of the Electricity (LCoE) produced with it. These items have been calculated independently with the aims to estimate both contributions as accurately as possible and to assess the driver of system economics.
The economic analysis of any technology that is still at a moderate TRL level is inherently uncertain and based on numerous estimates and forecasts which might become or not true. Therefore, it is mandatory to account for uncertainty and to assess how much the final results might be impacted by deviations in the input data set. This was done in two different ways, through the direct parameterisation of the results (i.e. by changing the input variables in a certain range) and with a simple statistical analysis based on the Monte Carlo approach. The latter is used to implement uncertainty in the market analysis whilst the former is employed in the direct estimation of costs.
The analysis described in the previous paragraphs was carried out for two business cases, namely the generation of electric power only (Business Case #0) and the simultaneous generation of heat and power (Business Case #1). Each business case is then divided into two possible operational modes: solar-only, where the system is run with solar energy only, and hybrid, where fossil fuel is burnt occasionally to make up for the lack of solar energy during cloudy periods of time or at night.
From a technical standpoint, it is also considered that the OMSoP technology can be implemented with moderate specifications and more high-performing cycle parameters. Two figures of merit are used to identify the standard and advanced systems: turbine inlet temperature (TIT) and recuperator effectiveness. The resulting combinations are shown in Table 6.1
The four cases shown in Table 6.1 are carried out through the entire analysis but, for the sake of clarity, most of the results are referred to just two of them. Indeed, the grey-shaded cells in the table stand for the reference or base-case system (800ºC and 85%) and the advanced system (900ºC and 90%). These are also referred to as Level I and Level II systems respectively.
Technology Levels I & II are initially implemented in a simple recuperative cycle with one compressor and turbine, a shown schematically in Figure 6.1. Nevertheless, there might be a value in adopting more complex layouts which can be implemented with essentially the same technology at the component level (Table 6.1). These complex systems are:
Intercooled recuperative system (ICR) whereby the compression process is split into two stages with an intercooling heat exchanger in between. The system still makes use of a recuperative heat exchanger.
Intercooled and reheat recuperative system (ICRR). In this case, the expansion process is also split in two expanders in series in between of which a second solar receiver is located.
There are three main items regarding the final cost of the electricity generated with OMSoP that must be supported by the end-user. These are presented below in chronological order as they have been developed in the project:
Installed cost of the technology. This is expressed in €/kWe. It accounts for the capital cost of the main equipment plus transportation and installation, including import/export duties.
Financial costs of the project, brought about by the operating costs and the finance-related costs like banking and debt pay-off. This contribution depends mainly on the boundary conditions under which the system is operated.
Yield. This is the amount of electricity generated which depends on the thermodynamic performance of the system and on the available solar resource at the system’s location.
A sketch outlining how the total installed cost is calculated is presented in Figure 6.2. It shows how the cost of the OMSoP system is broken down its main constituents to calculate the so-called Purchase Equipment Cost, which can be regarded in this context as the capital cost of the system ex-works (at the factory site). As evidenced, this cost depends on the layout and specifications of the system (see Table 6.1) and the manufacturing volume.
Component costs are calculated based on two main influential factors, one technical feature representing the size of the component and another functional dependence on the manufacturing volume as evidenced in Figure 6.2. The figure of merit for size is, of course, different for each component: mass flow rate for the gas turbine, diameter for the size, thermal output for the solar receiver and electric output for the Balance of Plant (BOP) components. Based on this approach, a complete literature review is performed to gather as much information about costs related to the OMSoP components. This means not only experience with dish and MGT integrated systems, like the work carried out by NASA in the 1980s, but also other systems using individual OMSoP components like dish-Stirling systems (cost of the dish), micro gas turbines and even solar receivers for process heat. The information obtained in this literature review is then completed with information provided by the OMSoP consortium members specifically.
The resulting database is then integrated and cost functions are developed for each component with the previously mentioned sensitivity to size and production volume. Some of these functions are shown in Figure 6.3 where the impact of the technology level is easily observed.
Further to this Purchase Equipment Cost, there are other cost items that must be added to the system: transportation, import fees and installation being the most relevant. These additional costs are driven by the location of the final system and they can largely influence the final installed cost as shown in Table 6.2.
The impact of location and technology/layout on the installed cost of the system is presented in Figure 6.4 for three locations: South Africa, Morocco and the East bank of China. The following nomenclature is used:
BC# stands for the business case: the code is 0 for electric power generation and 1 for combined heat and power.
SR / ICR / ICRR stands for the engine layout: simple recuperative (SR), intercooled recuperative (ICR) and intercooled and reheated recuperative (ICRR).
I / II stands for the technology level: Level I is for 800ºC turbine inlet temperature and 85% recuperator effectiveness and Level II for 900ºC and 90%.
The impact of location in figure 6.4 is evident. More favourable locations, i.e. locations with a more abundant solar resource, yield lower Installed Cost due to a smaller dish diameter or, reversely, a larger output for the same dish size. The final effect on cost can be as large as 10% if feasible locations are considered. Whilst this was to be expected, the impacts of technology level and engine layout are not so evident. Indeed, it is observed that there is a drastic drop in installed costs when higher components specifications are selected (upgrading from Level I to Level II). This is marked as (1) in the chart. On the contrary, the potential benefits coming from a more complex layout are not as large as initially expected. The transition from a simple recuperative system to a more complex engine using intercooling and reheat seem to have a much weaker impact on cost as deduced from the almost constant cost marked with (2) in Figure 6.4.
The information in the mentioned cost analysis is then integrated with the performance model developed to carry out annual simulations of the system in different locations. This is usual practice in project appraisal since the utilisation of annually averaged values of efficiency and insolation yield inaccurate results. The techno-economic performance of OMSoP in the selected locations is presented in Table 6.3 below with the same nomenclature used before.
There are several features in Table 3 that are worth noting:
The capacity factor (fraction of hours in a year that the system is in operation, referred to equivalent full load operation) and efficiency are strongly dependent upon location.
There is a large gain in efficiency when the system is upgraded from Level I to Level II technology.
When hybrid operation is considered (BC #0 – H), there is the potential to achieve 50% capacity factor with 90% solar share (minimum contribution from fossil fuels).
When Combined Heat and Power is considered, the electricity to heat ratio is raised with respect to other competing technologies like reciprocating engines. In particular, a 50% split is affordable with the standard Level II technology.
The same information is now presented in Table 4 for the intercooled configurations of the engine. The main observations when comparing the information with that in Table 6.3 are:
Annual efficiencies close to 20% are possible.
The electricity to heat ratio in CHP applications increases to over 70% due to the much higher solar-to-electricity efficiency.
The information in Table 3 and Table 4 translates into a Levelised Cost of Electricity as stated in the first paragraph of this section. This information is presented in Figure 6.5 where it is credited that the cost of electricity generated by OMSoP can be substantially lower than 10 c€/kWh in a good location and 10 and 15 c€/kWh in a mild one.
The information in Figure 6.5 becomes even more interesting when compared to the existing technologies in the market that are competing for the same applications and end-users. Such comparison is illustrated in Figure 6.6 showing that that:
OMSoP is more cost-competitive than dish-Stirling systems when performing the comparison for the same location and set of boundary conditions.
With respect to photovoltaic technology:
OMSoP exhibits the same performance as photovoltaic panels in a good location with moderately high direct normal irradiance (cases of South Africa and Morocco).
Photovoltaic technology is more cost-competitive in locations where the rated DNI is lower. This is due to this technology being able to exploit diffuse radiation.
These bullet points confirm that the initial objective of the OMSoP project to develop a Concentrated Solar Power technology to produce Solar Thermal Electricity that is cost-competitive against the existing technologies at the small-scale is accomplished.
This work was aimed at the screening and identification of potential markets that could be of primary interest for the commercialisation of the OMSoP technology. Such analysis is based on the combination of certain drivers influencing the likeliness that an area of the world becomes interested in the technology. These drivers are:
Availability of the solar resource (F_I), in particular in the form of Direct Normal Irradiance. Only areas with high DNI are expected to become interested in solar thermal technology.
Size of the eventual market (F_D), i.e. number of systems or total installed power that could be deployed in a particular country. This is directly related to the demand for electricity and, indirectly, to the population of the area.
Characteristics of the national grid (F_G). Countries where a fraction of the population does not have access to electricity are likely to become more interested in OMSoP.
Renewable energy policy (F_P). Countries with legal initiatives to foster renewable energies are more interesting as initial markets for OMSoP.
Financial risk of the country (F_F). As in any economic initiative, assuring the financial and market conditions of a country is fundamental if a new product is to be commercialised and capital is to be invested.
These factors are combined according to certain weights (which prioritise ones over the others depending on the cases considered) yielding the so called Index of Market Potential IMP.
A database with information compiled from scientific publications, national regulators, the World Bank database, Eurostats, the International Energy Agency and other institutions, to characterise these factors F_i as accurately as possible for a list of countries representative of all the areas of interest in the world. This information is then implemented into an Application whereby a flexible environment enables changing the weighting factors of the IMP for different boundary conditions. The graphical user interface is shown in Figure 6.7 where the different menus to change the weighting factors are visible as so is the option to assess an individual, stand-alone system or an array of several units (so-called farm arrangement) to produce hundreds of kilowatts.
A thorough analysis was performed to set a realistic/accurate set of boundary conditions because of which trustworthy information can be obtained. A sample of the results is presented in Figure 6.8 where temperature maps of IMP are shown for the stand-alone and farm arrangements along with the particular values for a list of selected countries. There are several observations that are worthy of note:
The markets of primary interest differ from one arrangement to another, in particular at the top of the list. The least interesting countries seem to be more or less the same in both cases.
Market stability and energy policies seem to be equally important to availability of solar resource and market size when it comes to the IMP of individual units.
Large countries with uneven distribution of the solar resource tend to be heavily burdened by the least favourable areas. Therefore, it is advised that these countries be divided into internal regions where the same methodology can applied with an input dataset that is specific to the region.
With regards to the most interesting countries for the future commercialisation of OMSoP, the analysis carried out revealed the following:
Countries like South Africa, Chile, Australia, United States, Morocco, Spain and Mexico seem to have a large potential for the OMSoP technology.
Countries with sparsely distributed population and high solar resources like Algeria or Morocco are also of interest for the stand-alone OMSoP system.
Countries with sunny regions that are lightly populated or large distance between regions with high availability of solar energy and regions that are densely populated are not as interesting; such is the case of Brazil or Italy. For these, large CSP facilities using other technologies are probably more cost effective.
Power plants made up of large arrays of OMSoP systems might find market opportunities in countries with high electricity demand and more favourable market conditions.
Uncertainty analysis is a transversal task which, from a practical standpoint, is directly implemented into the previously described cost and market analyses. The impact of uncertainty is assessed from two different angles. On one hand, a deterministic quantification of uncertainty can be implemented by merely performing sensitivity analysis with respect to the main input data set. This is used mostly in the cost analysis where the economic database compiles information from very different sources for each cost item which, in turn, can potentially take values within a range. How wide or narrow the range is depends on the particular item whose cost is being estimated.
Figure 6.9 illustrates this. The information in the charts shows the raw data that is used to produce the individual cost functions that were shown in Figure 6.3. It is easily observed that the different sources of information provide specific costs that are, in some cases, largely dissimilar. Such is the case of the micro gas turbine for which the lines in the upper plot are quite distant from one another, yielding important deviations in specific cost. On the contrary, the costs gathered for the receiver seem to be in good agreement.
The data in Figure 6.9 are weighted averaged considering the time value of money and the natural development of the technology with time. Then, a sensitivity analysis is performed to assess how much the final results in Figure 6.4 are affected by changes in the cost functions presented in Figure 6.3. The sensitivity analysis shows that the impact on the results is one order of magnitude lower than the main figures of merit, thus confirming that the results presented in the cost analysis are trustworthy.
The second angle from which uncertainty is approached refers to the market analysis. Indeed, market analysis is inherently uncertain because it takes into account future actions that might or not happen. Even if the current scenario were described and quantified very accurately (financial conditions for the investment, current number of end-users, and current demand of electricity in a country...) the future behaviour of the market would still hold a large degree of uncertainty; numerous examples can be found in handbooks covering this topic. For this reason, a more complex approach to uncertainty management is needed.
To manage uncertainty in the market analysis, the Monte Carlo method is used. This method relies on the same deterministic principle as the approach used in the cost analysis (see previous paragraph) but adds a statistical or random feature to it. Thus, the input variables do not take discrete values but, on the contrary, they take values in a range with a probability function assigned. When the impact of uncertainty is assessed, the same application described before is run thousands of times, each one of which sets up the input variable set with samples from the cited probability functions. If the number of runs is high enough, one can be confident that the whole range of values has been tested, even those with a lower probability, and the impact on the results can be evaluated.
The practical implementation has been made with standard probability functions and with other functions, yielding somewhat similar results. This is observed in Figure 6.10 where detailed information about the market analysis for India is presented. The statistical nature of the assessment is evident in the upper row of charts where the input variable values covered in the Monte Carlo analysis is presented. As shown, the analysis is run with a very wide of input data. The bottom row presents the position occupied by the country in a ranked list according to the countries’ IMP. The impact of varying some of the input factors F_i is strong (the position in the rank list changes) whereas other factors have a very weak effect.
The information presented graphically in Figure 6.10 can also be presented numerically. It is very interesting to see that even when large uncertainty is added to the input dataset, the resulting list of countries ranked according to the Index of Market Potential remains quite unaltered with just minor changes one or maximum two positions up or down.
Potential Impact:
The OMSoP project has demonstrated the technical and economic feasibility of integrating a concentrated solar power parabolic solar dish system with a micro gas turbine to generate electricity. The electrical power output from a single unit with one parabolic would be in the region of few kW to about 25 kW. This power range for a single unit is suitable for small to medium scale domestic, industrial and public services applications. The units are modular and hence can be stacked to produce larger power outputs if needed, potentially in the region of Megawatts to tens of Megawatts covering a gap in the available solar thermal power technology which is only economic for much larger power outputs. In comparison to the popular photovoltaic systems, the OMSoP system can be transferred to a fully dispatchable power plant through hybridisation with fuel combustion or thermal energy storage, or possibly both in the same units to achieve reliable source of electricity. The most suitable locations in the first instance would be off-grid locations or those with low grid reliability. This however does not preclude the system from direct competition with other renewable energy systems connected to the gird as the techno-economic studies have shown.
At the end of the project, it can be claimed that the proposed OMSoP technology has achieved a technology readiness level between 6 and 7 where a prototype was tested in an environment very close to operational one.
The project has addressed three key impact areas through the dedicated technical work packages. These are:
1. System component development (Work Package 1), were system components were developed: receiver, concentrator and MGT to provide technical solutions to issues that have previously prevented the wide spread deployment of similar systems. For instance, a high efficiency micro gas turbine was developed that would lead to reduction in dish size which is the single most expensive part of the system.
2. System design and integration (Work Package 2), aimed at integrating the components developed in WP1 into a functioning system by system design and optimisation, also aimed at demonstrating the system which provides a proof of concept and feedback for component optimisation. The outcome of this activity has been extensive knowledge gained in the field of solar powered micro gas turbines in terms of sourcing of components, manufacture, integration, electrical scheme, control system and many other detailed aspects that would provide clear direction for future work towards commercialisation.
3. Techno-Economic analysis (Work Package 3), aimed at developing thermodynamic and mechanical models of the system and its components, as well as further insight into future deployment by studying concepts such as medium and long term storage hybridisation with other fuel and MGT power argumentation. Another objective has been to identify the potential markets for such systems worldwide just paving the way for targets for initial deployment.
The ultimate impact of OMSoP is to introduce a step change in the current paradigm of Concentrated Solar power technology which, if the current techno-economic scenario remains, will progressively be less competitive compared to PV and wind. The OMSoP concept, which aims to widen the scope of solar thermal technology from the current emphasis to centralized generation towards distributed generation targeting small and medium sized applications. This will tackle two major hurdles in the current solar power market. On one end, compared to centralized generation, it will reduce significantly the cost of the grid infrastructure particularly in countries with many small scale scattered communities. This will also reduce grid losses and allow for combined heat and power, combined cooling and power or multiple generation at the point of consumption hence increasing the overall efficiency of the system. On the other hand, and compared to PV used in distributed generation, the OMSoP system has the advantage of flexibility to be extended to hybrid systems with co-firing and/or thermal energy storage allowing for potential full dispatchability, compared to PV which needs either another system such as diesel engines or battery storage, the cost and environmental impact will be significantly lower.
The successful completion deployment of the demonstrated OMSoP technology thus could contribute to a new scenario of the Energy sector in several regions in Europe and many other countries with high solar irradiation wherein this type of energy technology will ensure dispatchability and cost competitiveness simultaneously. This has not taken place yet since PV and wind are cost-competitive but not dispatchable whilst state-of-the-art CSP is dispatchable but has relatively very high costs and only suitable for centralised power plants. Indeed, dispatchability sets CSP apart from other technologies with regards to increasing renewable energy penetration. This has been confirmed by a study carried out by IEA and IRENA (Renewable Energy and Jobs. Annual review 2017. IRENA) where it is claimed that the overall installed capacity in CSP will increase from the actual 4 GW, to 100 GW by 2030 and 300 GW by 2050, resulting in a yearly installed capacity of 10 GW in 2030 and 20 GW in 2050. The assessment of the cost of electricity from OMSoP has been shown to be competitive with PV and conventional power generation through the studies conducted in work package t3 of this project. This suggests that if the right interest by industry and investors is achieved through appropriate dissemination of the outcomes of the project, it is more than likely that the proposed system will make its way towards commercialisation.
The following list evidences how OMSoP has addressed the expected specific impacts put forward in the European Comission work programmes
• Reduce the technological risks for the next development stages: This project has demonstrated that the proposed technology is feasible from both rechincal and economic points of view. For the first time worldwide a micro gas turbine of the proposed power range was successfully tested operating with concentrated solar power using a single parabolic dish concentrator. All the gained knowledge on concentrator, receiver, micro gas turbine, electrical eclectronic and control componentes, materials and component manufacturing process has provided new knowledge and experidence that will reduce the technological risk for the future comercialisation activity of this technology.
• Significantly increased technology performance: the utilization of a micro gas turbine parabolic dish concentrated solar tower technology for distributed and small scale centeralise power has shown to yield competitive technologies in terms of cost and reliability. The gain is expected to be higher if the cost of land is taken into account where the proposed system requires about half of that needed by PV for the same power output.
• Reducing the life-cycle environmental impact: the deployment of OMSoP will contribute to the widespread deployment of CSP technology which is renewable and with negligible CO2 footprint. Compared to other solar power technologies, OMSoP will further reduce its life-cycle enviromental impact as a result of the smaller power block, therefore requiring less materials, and most importantly, lower land use due to the smaller solar field.
• Nurturing the development of the industrial capacity to produce components and systems and opening of new opportunities: Industrial partners will be actively involved in the development of OMSoP technology for future deployment. Further European industry is expected to join future commercialisation activity including those from Micro gas turbine and solar dish manufacturers.
• Contributing to the strengthening the European industrial technology base, thereby creating growth and jobs in Europe: the development of an innovative CSP technology will take Europe to the forefront of the renewable energy industry. Indeed, European industries have led CSP since the early 2000’s. In order to strengthen the position of the European industry and ensure competitiveness against the fast-growing Far East industry, the development of highly innovative concepts must become a priority in the agenda of European decision-makers. The unfortunate experience in the PV industry, where most of EU manufacturers closed down when challenged by the Chinese industry, must be born in mind in this respect. Certainly, the project presents a groundbreaking concepts successfully demonstrated and also innovative methodologies that would give an edge to the European industry.
• Reducing renewable energy technologies installation time and cost and/or operational costs, hence easing the deployment of renewable energy sources within the energy mix: the cost of electricity reduction will be achieved through the higher overall conversion effiency, when taking into account utilisaiton of the dwonstream heat resulting even more competitve technology.
• Increasing the reliability and lifetime while decreasing operation and maintenance costs, hence creating new business opportunities: the OMSoP technology will achieve these targets through the more compact and reliable micro turbine compated to Striling engines. The other markets that can take advantage of the OMSoP technology. Additionally, micro gas turbines can be used for many other applications such as CHP, auromotive range exteders and back up power systems resulting in mass production and reduction in unit price.
• Contributing to solving the global climate and energy challenges: the OMSoP consortium is very concerned to the energy-related environmental problems that will be faced by future generations. This is why the technology exhibits numerous features oriented to solving or at least mitigating these problems: (i) OMSoP aims at contributing to a renewable-energy dominated world through the worldwide deployment of Concentrated Solar Power, (ii) within the outcome of the project, a dedicated studies covered the environmental performance of the technology, including Life Cycle impact and Greenhouse Gas emission reduction; (iii) a key objective of the project has been the simultaneous optimization of the overall solar plant efficiency in order to allow an LCoE reduction with respect to other technologies, therefore making solar electricity affordable including the case for developing regions of the world (key aspect of the energy challenge since environmentally clean electricity is still generally more expensive than electricity from fossil fuels).
• Other impacts OMSoP would ultimately lead to a fully dispatchable Concentrated Solar Power technology that is cost-competitive against any renewable energy technology and, as such, it will contribute to the diversification of energy sources in Europe. Additionally, due to the hybrid operation capability, it can partly or entirely replace conventional power plants. It must be noted that not only can the OMSoP concept be used in Concentrated Solar Power applications, but it can also be employed with other renewable energy sources such as biomass in co-firing or thermal energy storage.
Impact on developing countries and regions:
The electricity distribution networks are under developed in many parts of the world. In these areas: power grid network structure of high-voltage substation are very weak and their supply is not reliable; there is a lack of sub-stations, which causes each station’s coverage area to be too large and main substations are regularly out of repair; supply area of each site is not clear and power lines can be in chaos and power losses are typically in addition to mismatch between high voltage main lines and branch lines leading to unreliable supply. It can be inferred that, electricity supply cannot be guaranteed in poor areas, that is, frequent power outages and chronic power outages are common; voltage is low and no power supply during peak times. There are several disadvantages for off-grid household PV system complemented by wind; firstly, power availability is climate dependent. Additionally, low power density which would occupy a large area for higher capacity power system leading to high initial cost estimated at 3~15 times higher than conventional power generation systems. Secondly, large-scale application of wind-solar complementary power system is limited by high cost of energy storage systems if around the clock power is to be provided. High cost of pumped-storage systems which are also cite specific, electrochemical energy storage like lead-acid battery storage and their relatively short lifetime makes them economically not viable. On the other hand, many of these areas have high solar energy resources, but these can be with no electricity supply or insufficient power supply. Commercialisation of the OMSoP technology in particular, when coupled to thermal energy storage can lead to transformations in such societies at different levels. The impact to local communities in the following is conceivable, sometimes interdependent in relation to providing sustainable development, reducing poverty, improving the quality of life and health provision as follows: (1) the provision of clean reliable and affordable source of electricity using an abundant and sustainable resource. (2) The provision of local jobs through employment in the manufacture of system components, assembly, installation, operation and maintenance in addition to satellite jobs in the supply chain and services. This would help to improve the welfare of the local populations and reduce poverty. (3) Freeing up populations time from fetching primitive energy resources or coping with unreliable energy, which would allow more time for education and training providing better jobs and higher pay. (4) Availability of electricity for services, in particular health, would improve health, raise life expectancy and improve quality of life for people and reduce the cost of health services to local governments freeing up resources for further development.
Dissemination
To maximise the impact of the OMSoP project, the team has embarked on a wide range of dissemination activities to explain the project and increase future interest from potential industries, investors as well as customers. The following dissemination activities were undertaken:
• Project website (https://etn.global/research-innovation/projects/omsop/): The project web site was made live at the start of the project and will remain active for several years after the end of the project. It contains information about the technology, its operating principles and advantages as well as a description of the project structure. All project public reports are also made available on the website of access of the public.
• A trifold leaflet distributed to tradeshows and exhibitions (e.g. ASME Expo Turbo (2014 Dusseldorf, Germany, 2015 Montreal, Canada), POWERGEN (2014 Cologne, Germany; 2015 Amsterdam Netherlands), International Gas Turbine Conference (Brussels, Belgium), ETN’s Annual General Meetings in 2014 and 2015, as well as at all ETN meetings in Brussels.
• A presentation of the OMSoP Project and results at ETN’s AGM in Paris (2014) and in Dublin (2015).
• A presentation of the OMSOP project at the National Research Centre for Gas Turbine and IGCC Technology of China, January 2015, Beijing, China
• OMSOP project slides included in ETN’s general power-point presentation disseminated at various meetings and events globally (2013-2017).
• OMSOP project information and links on ETN website, in ETN’s Brochure and ETN’s Members package.
• A bi-monthly electronic update sent to the consortium with the status updates, the project reports and meeting information.
• Two video describing how the OMSoP system works, the objectives and the result achieved in the project and the challenges overcome in the project by the consortium partners.
• A press release of the OMSoP project highlighting the objectives achieved.
•
• Publications of articles on the worldwide press (e.g. Gas-To-Power Journal, Gas Turbine World, etc.).
• ETN continuously disseminates the OMSoP project progress and results through the ETN network via its website, Twitter, LinkedIn, Quarterly Newsletter and at the Annual General Meeting.
List of Websites:
https://etn.global/research-innovation/projects/omsop/
This project aims at the development and demonstration of concentrated solar power (CSP) parabolic dish system that generates electricity using a micro gas turbine (MGT). The technological objectives are to identify the components and overall system design requirements for producing an efficient reliable and cost effective system that can generate about 3 - 25 kW of electricity per unit using a single dish. The vision is to have modular systems that can be used either for distributed power generation or to be stacked in a farm arrangement to form medium size solar thermal power plants of several MWs of power, a range which is not economical if the commonly used steam turbine based CSP power plants are used. Parabolic dish CSP technology has been under development in recent years using Stirling engines as the power unit. However, questions have been raised about the reliability of these engines as well as their suitability for hybridisation with other fuels and thermal energy storage. Hence the unique feature of this project is the introduction of the micro gas turbine as the prime mover. Micro gas turbines have the potential to be lighter, more reliable, cost effective and flexible for hybridisation as well as the utilisation of exhaust heat in multiple generation maximising the overall efficiency of the system. The development of the demonstration plant incorporating a 6kWe micro gas turbine has been accompanied by the design and laboratory testing of short term thermal energy storage to smooth out fluctuations to variable solar irradiation. Additionally, techno-economic and worldwide market analyses have been conducted to identify the most suitable markets for the technology as well as the overall cost of the system upon volume production. Furthermore, system optimisation studies as well as life cycle analysis were performed to guide future developments of system.
The OMSoP project consortium is composed of eight partner organisations from five European countries from research organisations and SMEs with track record of research and project management in related areas. The project was organised into three technical work packages and a management work package led by City University of London with Professor Abdulnaser Sayma as the project coordinator. The technical work packages (WPs) are component development WP lead by KTH, system integration and demonstration WP led by ENEA and techno economic optimisation lead by RO3. The remaining project partners were INNOVA, responsible for concentrator design and installation, CP participating in the micro gas turbine control system and overall system optimisation work, USE working on market and cost analyses and European Turbine Network working on dissemination activities.
The project required 52 months to execute and resulted a successful demonstration of the solar powered micro gas turbine technology where the system was tested in Cassicia at the site of the Italian National Agency for New Technologies, Energy and Sustainable Economic Development (ENEA). Laboratory tests on the short-term energy storage demonstrated the capability of allowing smooth operation of the system. System optimisation studies produced several alternative arrangements for the power plants for future systems. The market and cost analysis showed that the OMSoP technology could be competitive with other solar power technologies in regions with high solar irradiation and favourable technological and economic conditions. The main conclusion from this project is that parabolic dish micro gas turbine technology is promising to be feasible, cost effective and reliable alternative to existing solar technologies and could open new markets. A framework should be put forward to develop the technology towards commercialisation.
Project Context and Objectives:
The main concept of generating electricity from solar insolation using a thermal power conversion unit is based on concentrating sun rays at a focal point where the temperature attainable rises to an adequate level for conversion to mechanical power. A working fluid, such as air, circulating through a prime mover that converts thermal power to mechanical power and then electricity, is heated at the focal point. Thus, a typical CSP system is made of a concentrator, a receiver-absorber and a prime mover. The concentrator is typically made of curved mirrors to concentrate solar radiation as much as possible and hence achieve high working temperatures. The receiver-absorber is a device located at the focal point which continuously extracts the thermal energy from the solar radiation. Part of the solar energy impinging on the receiver is transferred to one or more working fluids, while some of that energy is lost to the surroundings. Current small scale plants based on parabolic dish concentrators rely on Stirling engines as the prime mover or on concentrated PV. However, Stirling engines are known to have numerous technological problems affecting their reliability.
The utilisation of solar energy for power generation represents an important element of the renewable energy mix required to meet the environmental constrains in a suitable manner. Several technologies are currently being used to harvest the abundant solar energy such as photovoltaics and solar thermal power plants. Both technologies have made big strides in recent years in improving performance and in cost reduction. Parabolic trough collectors, or solar towers in conjunction with steam turbines are suitable, from practical and economic perspectives, for large scale power generation in the order of at least tens of Megawatts. Photovoltaics has been systematically reducing in price and are being deployed at scales from domestic applications to the Gigawatt scale plants, but back up power and energy storage remain major challenges for further deployment of such systems. A less developed solar thermal technology suitable for distributed power generation and power plants in the order of Megawatts is the parabolic dish plants that use a single dish to concentrate solar power at the focal point where the temperature of a compressed working fluid in a prime mover can be raised and the fluid is expanded to produce power. Both Stirling engines and micro gas turbines lend themselves as candidates as a prime mover for such systems. Parabolic dish systems using Stirling engines have been under research, development and demonstration for several years, however, their market deployment did not live to the original expectations due to several factors such as reliability and economy. Micro gas turbines as a prime mover in parabolic dish systems were trialled as early as the 1980s, but the unfavourable market conditions for renewables at the time were probably the main reasons that such technology was not taken further. Other recent attempts to integrate micro gas turbines with parabolic dish systems attempted to utilise of-the-shelf micro gas turbines not specifically designed for the purpose and were also much larger than could be sustained by a single dish leading to limited success. This did not lead to further developments. The Optimised Micro Turbine Solar Power (OMSoP) project was conceived to develop and demonstrate a purpose built micro gas turbine utilising a parabolic dish aiming to pave the way for successful market uptake due to realisation by the project team that specific features of such technology will make it suitable for applications and regions of the world more than other solar power technologies. This would be enabled by developing a reliable and efficient system with competitive cost of electricity in the target markets and application areas.
The vision of the OMSoP project has been to develop innovative concepts that can improve the ratio of the electric power generated to the solar energy insolation on the dish and to improve operability in relation to solar energy short time fluctuations through the development of short term thermal storage built within the receiver. These are important features that are expected to drive down installation and operation costs with respect to other solar technologies that should be accurately identified. Ultimately, the vision is that this should result in a solar dish-MGT system which is competitive to Photovoltaic (PV) and solar dish-Stirling engine systems. Subsequently, the aim has been to carry out a successful demonstration of a modular solar dish-MGT power plant which could in, a second phase, be deployed and taken up by the market in countries with high direct solar irradiation, such as those bordering the Mediterranean as well as the vast export markets such as China where the system is anticipated to be suitable to remote areas with no cost-effective provision of centralised power plants.
The minimum environmental impact of concentrated solar power (CSP) systems brings about the possibility to generate power close to the consumers, hence benefitting from eliminating transmission losses as opposed to other fossil fuel power stations. Due to the variation of demand depending on population density and the variable typical energy requirements per capita, a modular system which can be stacked to produce the local power requirements will be an ideal solution for rural and farming communities, but it can also be stacked in larger numbers to provide medium size power demand in the order of Megawatts.
The aim of the OMSoP project has been the development of plants that take current technology beyond the state-of the-art by integrating a highly efficient concentrator and receiver-absorber. The Stirling engine has been replaced with a micro gas turbine (MGT) with efficient power cycle which can be arranged with sub-cycles adequate for the power levels of the target application. The aim has been to develop plants that have performance and cost advantages that make them competitive with dish-Stirling and also with conventional fuel plants in certain markets; they also have the potential to compete with PV on dispatchability when integrated with thermal storage and/or co-firing using conventional or bio-fuels. Ultimately, this should pave the way for their widespread deployment in the future as a viable, reliable and cost effective clean energy source.
The ultimate goal of the OMSoP project has been to provide and demonstrate technical solutions which will allow future deployment of small scale reliable and cost effective concentrated solar power dish systems to enable industry to invest in the next generation of these devices. To achieve these objectives, it was required to address technological challenges with the primary components, the concentrator, the receiver-absorber and prime mover, leading to a demonstration of the concept.
In the concentrator, mirrors, made of plated glass or other reflective materials, typically contribute to the high initial cost of the solar dish systems, and, even if they have a high reliability, they can be too heavy, requiring robust positioning mechanisms and support structures. The primary challenge for the parabolic dish concentrator is the development of a new mirror technology that can ensure high performance at a competitive cost and an easily transportable size.
The main challenges for the receiver-absorber within the parabolic dish system were reliability and cost. Existing designs of the receiver make it one of the life-limiting components in the entire system. This is due to thermal stresses, thermal cycling, the use of brittle materials for absorption such as ceramic foams and the deteriorating efficiency of the sealing glass window under high thermal loads. Localised overheating and flow instabilities can also occur and these also serve to limit the effective life. Other challenges result from the high temperature that may result re-radiation of energy and significant losses to the surroundings. The aim here is to systematically address those challenges to meet the overall system performance, reliability and cost requirements.
A primary feature of this research is to replace Stirling engines, typically used in the target power range, with MGT. Gas turbines are known to have good weight to power ratio, reliability and ease of maintenance. However, downsizing existing MGTs to a size suitable for a power rating less than 10 kWe brings about several major challenges. The high rotational speeds required, typically in the order of 140,000 rpm, put significant pressure on the bearing design and rotordynamic stability, the latter is compounded by the need to the system to have wider operational range than conventional MGTs due to the variable solar irradiation. Other challenges are associated with the high turbine inlet temperature and reliable, light weight recuperator design, needed to achieve high MGT efficiency, and thus reducing the overall size of the concentrator. Such aspects have been addressed in this research.
Further studies aimed at the development of short term thermal storage technology that could be integrated with the receiver to smooth out short term solar irradiation fluctuations. Optimisation studies also aimed at identifying potential cycle arrangements and multiple generation concepts for maximum overall efficiency.
To enable commercial success, it has been also important to conduct detailed techno-economic study of the proposed system as well as to perform worldwide market analysis to identify the most suitable markets for future deployment. Both studies aimed to paving the way for future phases of the technology development toward full scale commercialisation. The target system has been expected to achieve the following outcomes:
- A high efficiency solar concentrator and receiver system that may include integrated short term storage capability and fossil fuel backup.
- Reliable micro-turbine system.
- Stand-alone machine.
- Modular, lighter and more compact system with simple cooling requirements.
- Low overall cost, easy installation and reduced maintenance.
- Potential for integration with medium and long term storage.
A life-Cycle Assessment will be performed including requirements for standardisation and integration with the European policy and legal framework, for economical assessment and market exploitation.
Project Results:
To achieve the OMSoP project objectives, a work plan has been designed which required the organisation of the project into three technical work packages (WPs) as follows:
Work Package 1: System component development, this included development of a solar concentrator, development of a solar receiver and development of a micro gas turbine. This work package including the following partners:
Kungliga Tekniska Högskolan (KTH) (WP leader)
City University of London (CITY)
INNOVA Solar Energy Srl (Innova)
Compower AB (CP)
University of Seville (USE)
Univertiy of Roma TRE (RO3)
Work package 2: System design and integration, this included overall system design, integration and testing representing the demonstration activity of the system in representative conditions. This work package included the following partners:
Italian National Agency for New Technologies, Energy and Sustainable Economic Development (ENEA) (WP leader)
CITY
RO3
CP
KTH
USE
Work package 3: Techno-Economic analysis, the work focused on two major aspects, system optimisation considering several alternative arrangements and market and cost analysis for future deployment. This work package included the following partners:
University of Roma TRE (RO3) (WP leader)
CITY
ENEA
CP
KTH
USE
The scientific and technological achievements of the project will be grouped below under the following categories: (1) the solar concentrator, (2) the solar receiver, (3) the micro gas turbine, (4) system integration and testing, (5) system optimisation and (6) market and cost analysis. More details can be found in the project deliverables available in the project web site (https://etn.global/research-innovation/projects/omsop/).
The solar concentrator
INNOVA conducted a study to identify the optimum arrangement for a concentraed solar power parabolic dish system arrangement. It was found out that a single dish with the power unit installed on the arm is the most appropriate for the target power range and for stacking several systems to form a power plant. The conducted studies on the dish structure and mirror materials suitable for the demonstration plant and finally performed a design of the dish and sun tracking system. The assembly of the dish was completed in April 2015 while the experimental characterization of the dish on-site started in May 2015. The dish characterisation was accomplished by ENEA trough two different methods: A) via simulation, mainly considering the experimental data of shape and solar reflectance of the facets composing the dish (indirect method); B) by direct experimental measurement.
The reflectance measurement of the Almirr mirror, which is the reflecting material selected by INNOVA, was performed at the ENEA labs.
The dish shape evaluation required two additional experimental steps: geometrical characterisation and facet canting. The geometrical characterisation of the dish structure was accomplished through the photogrammetry method.
The facet canting (mirrors orientation) procedure was carried out adopting a new experimental method developed by ENEA (Visual Inspection System method)
The facet canting took a couple of days; Figure 1.1 shows the progress from the beginning to the conclusion of the procedure. At the end of the canting a large part of the dish area appeared red-coloured (aligned); the remaining part is essentially black, meaning imperfection of the mirror shape.
Then the dish shape measurement was performed trough the VIS method to evaluate the slope deviation of the dish surface from the ideal one. At that purpose an innovative experimental set-up named VISdish was built up.
The outcome of the dish shape measurement is reported in Figure 1.2 where the slope deviation, peaked at 5-meter radius, is represented; from these results the limited optical quality of the dish was clearly noticed, most probably due to mirrors shape deformation during the on-site installation.
Based on the experimental data collected (mirrors solar reflectance and slope deviation), an accurate simulation of the concentrated flux around the focal point of the dish (Figure 1.3) was performed through the application of the SIMULDISH optical model, developed by ENEA. The dish optical efficiency detected was lower than the initial prediction and, accordingly, a resizing (enlargement) of the solar receiver was performed through the application of the developed system model to keep a constant power input to the MGT.
To validate the results obtained through the application of the SIMULDISH software (indirect method), the concentrated solar flux distribution was experimentally measured using an optical target positioned in correspondence of the focal point (direct method “Flux Mapper”, Figure 1.4).
The experimental tests were initially carried out during the night (moon tracking, Figure 1.5) to work in safe condition and get familiar with the instrumentation and subsequently were replicated on-sun, whose results are represented in Figure 1.6.
The outcome of the direct evaluation method was quite consistent with the indirect simulation, even if a worse dish performance was detected in the first case
A theoretical analysis was performed by ENEA to resize and propose new solutions for the realization of a new prototype of a solar parabolic dish collector to be integrated with the receiver/micro gas turbine (MGT) system developed within the OMSoP project.
Two new possible configurations were defined, starting from a critical analysis of the results obtained on the prototype installed at the ENEA Casaccia Centre. As a first option, it was assumed to have a system with the same electrical power output as the prototype (7 kWe), while in a second option a higher electrical power output (15 kWe) was assumed. Figure 1.7 shows a sample of the results.
The solar receiver
Two receiver concepts have been developed for risk mitigation: a robust cavity and a volumetric receiver offering a higher efficiency. For the demonstrator unit the more robust but less efficient cavity receiver has been chosen. Both receiver types were tested in the KTH high-flux simulator (HFSS) and based on the testing results the concepts were optimized for the use in the OMSoP solar dish demonstrator. This testing includes receiver component test as well as integrated testing of the receiver with a micro gas-turbine (MGT). Finally, the optimized cavity receiver was tested in the OMSoP demonstrator. RO3 has investigated the Short Term High Temperature Thermal Storage Receiver (STHTTSR) concept development.
For the demonstration plant a receiver concept that uses impinging jets to enhance the heat transfer was chosen. It was designed using an inverse design method as well as conjugate heat transfer analyses. Initial receiver component tests in the KTH solar simulator showed that the concept is promising for the integration in the OMSoP system. It was found that it is important that the absorber walls are fully oxidized before the receiver can be installed and operated in the real dish system. In a second step the cavity receiver was coupled to a MGT and the KTH solar simulator simultaneously to investigate the dynamic behaviour of the coupled system as well as to obtain improved measurement results of the cavity receiver performance. It was shown that the impinging receiver is suitable for the integration in the OMSoP solar dish system. During the experimental campaign air outlet temperatures of up to 810°C were obtained with absorber cavity material temperatures at around 960°C. The mechanical and life objectives were met with this design. It was shown that the total energy loss of the receiver is around 9.15 kW, and the efficiency of the receiver can be estimated to reach 73.7% at its design boundary conditions.
A receiver couples with short term thermal storage was developed and tested in the laboratories of RO3. The idea was to smooth out short term fluctuations of solar irradiation including the passing of clouds for periods of several minutes. The short-term Heat Storage Receiver (StHSR) concept shown in Figure 2.1 is based on a tube bundle system immersed in a Phase Changing Material (PCM). The container has a front plate that is directly irradiated by the concentrated solar energy. The heat is transferred into the PCM by means of the back of the frontal plate and solid heat pipes (acting as fins), thermally connected with the frontal plate and immersed in the PCM that provides for the heat transfer to the working fluid. During the heat storage, the liquid phase fraction increases because the entering heat power is higher than the exit one. When, due to the DNI fluctuations and partial or total solar eclipses, the entering heat power becomes lower than the exit one, because of the latent heat released by the PCM solid fraction increase, the working fluid exit temperature variations are damped in respect to that caused by direct irradiation of the tube bundle.
The preliminary design has led to select a tube bundle system made of 12 tube bundles to be connected by an external muffler as shown in Figure 2.2. This solution has been chosen to use the same receiver for complex multi-shaft MGTs with multiple heating to achieve high efficiencies of the same order of the Stirling engines for CSP applications.
Since the level of knowledge on such kind of receiver is almost non-existent, a theoretical model based on a FV approach with lumped phenomena in the volume nodes has been developed to allow for implementation of the not yet known phenomena that would be evidenced by the experimental activity.
The short-term heat storage receiver has required tests on models to understand how the melting and the solidification occur inside the volume and to understand the heat transfer process. The models have been tested in the RO3 solar lab using a spiral tube bundle (Figure 2.3) where xenon lamps, reflectors and Fresnel lens have been used to simulate the concentrated sunrays. Figure 2.4 shows a view of the test facility. Tests were carried out by filling the container by a NaNO3/KNO3 60/40 salt provided by ENEA with the front plate having a Smooth back Surface (case SS) and equipped with solid heat pipes (SHP). Results showed that the SHP increase strongly the heat transfer rate. The frontal surface thermal temperature has been measured by a pyrometer. Further tests have been carried out by measurements made by thermocouples immersed in the charge. The front plate temperature distribution has been measured also by a thermos-camera. The hot spot and the temperature distribution along one plate diameter are shown in Figure 2.6. The temperature measurements versus time are shown in Figure 2.7. These experiments allowed the calculation of the power impinging entering the frontal surface of the plate and the emissivity of the same surface. Moreover, the heat power transferred from one volume to the other has been evaluated.
A Finite Volume Model based on lumped parameter concept, sketched in Figure 2.8 has been developed for simulation purposes. The model is flexible because various phenomena can be activated according to the experimental evidence. Simulations reported in Figure 2.9 have evidenced similarity in the front plate tested temperature shown in Figure 2.6 and great capability of the receiver to soften the oscillations induced by a DNI step change and sinusoidal oscillations. An example is given in Figure 2.10.
Different tube bundles have been evaluated. Such tube bundles were arranged as spirals, U tubes and Helix Bundles. The last ones have demonstrated the best behaviour and then the final arrangement has been sized accordingly. Twelve tube bundles are required to cover, by parallel and series multiple connections, various MGT arrangements that, in the perspective, they can be used to satisfy market requirements. LiF has been selected as PCM for 800°C MGT TIT and MgSi has been selected for 900°C TIT. The weight of the LiF PCM in the receiver to supply heat power for 45 minute with zero DNI is about 43kg. Longer time interval of full heat power supply by the receiver increases the receiver weight that requires it would be installed on the ground.
The performance of the cavity receiver was evaluated in the KTH HFSS. Figure 2.11 shows the front (A) and side (B) view of the solar receiver installed in the HFSS.
To calculate the receiver performance indicators pressure and temperature measurement have been taken at different positions. In Figure 2.11 can be seen pressure transducers (PT), K-type thermocouples (KTC), and an IR-camera. Compressed air at approximately 20°C enters the receiver at the side through a measurement pipe which is fitted with thermocouples and pressure transducers. The mass flow can be adjusted remotely using a mass flow controller upstream. Inside the receiver, the air is heated by concentrated radiation supplied by the solar simulator before expanding in the choked nozzle which connects to the ventilation system using a flexible high temperature hose.
Initial tests of the cavity receiver were performed without integration of a MGT and without the electric air heater which means that the air entering the receiver was approximately 20°C. First, the non-oxidized receiver was gradually warmed by 4 lamps, 8 lamps and finally 12 lamps with an air mass flow input of 30g/s at an absolute pressure of 3 bar. After one hour of testing under the full power of 12 lamps, it was observed that the bottom region of the receiver was significantly oxidized which happens at a temperature level above 800 °C, as shown in Figure 2.12 even though the thermocouple on the cavity wall showed that the temperature on the cylindrical surface only reached approximately 340°C. It was concluded that more radiative energy is absorbed in the bottom region of the cavity as compared to what it was designed for.
As the cavity receiver was designed to operate at temperatures of 800°C and higher, the cavity material temperature reaches values above 800°C which means that the material (especially metallic material) surface would be fully oxidized. During the design process, the emissivity of the material with a fully oxidized surface was used instead of the emissivity of the material with a polished surface. For a receiver with a non-oxidized absorber surface the radiative boundary condition could be totally different to the design conditions. Normally, an oxidized surface has a higher emissivity than the metal itself. Therefore, the high flux region on a non-oxidized absorber is located further towards the aperture than on a same sized oxidized absorber due to the multi-reflection inside the cavity. The bottom of the non-oxidized absorber, where no impinging jets are distributed, might work under an undesirable high light flux level and overheat condition. Moreover, the cylindrical wall region located under the impinging cooling jets works at a lower temperature than the design temperature and might work below the oxidation temperature. This phenomenon might cause receiver failure if it would be fully focused without pre-oxidation directly after installation on a real dish system. The results imply that for using the impinging solar cavity receiver the real dish system it is imperative that the absorber is fully oxidized before installation.
Before the following detailed testing, a pre-oxidation process of the absorber was attempted by using an Acetylene burner. After the oxidation process, the cavity absorber is closer to a black body than the cavity before testing and the cavity after one hour testing with 12 lamps, as shown in Figure 2.12 (C). However, a full oxidation could still not be accomplished on the side wall surface which means that the optical efficiency is reduced due to higher reflectivity on the side walls.
The pressure drop across the solar receiver remains well below 0.1% for the operating points investigated. It can be seen that the total efficiency decreases significantly from 53.0 % to 42.9 % with the decrease of the mass flow from 30 g/s to 15 g/s. Due to the decrease of the mass flow, the heat convection efficiency on the back-side of the receiver also decreases. This in turn leads to the increase in wall temperature as seen in the experimental measurements of the wall temperature (the larger value measured by the two thermocouples on the wall). It increases from 329.6 °C to 435.2 °C with the decrease of the mass flow from 30 g/s to 15 g/s. Therefore, both the radiation and convection heat losses increase and finally cause the total efficiency to decrease. At nominal mass flow operation and design radiative boundary conditions the efficiency is calculated to be 73.7%.
Analysing the test, it can be concluded that the cavity receiver concept using impingement jets is promising for the integration in the OMSoP system. It is important though that the absorber walls are fully oxidized before the receiver can be installed and operated in the real dish system.
In parallel to the design and testing of the cavity receiver a volumetric receiver concept was designed and a prototype built for experimental evaluation in the KTH HFSS. To obtain representative results when testing the volumetric receiver in the solar simulator scaling effects were investigated between a receiver installed in the OMSoP solar dish and the KTH HFSS. This study is based on a coupled CFD/FEM analysis in combination with ray-tracing routines that have been validated against measurement data of the OMSoP solar dish and the solar simulator. Because of the analysis, a scaled prototype was manufactured.
The receiver was placed in the focal spot of the HFSS and connected to a high-temperature air heater at the inlet and a chocked nozzle to simulate the MGT at the exit as shown in Figure 2.13. Compressed air is supplied by a high-pressure compressor and the mass flow is adjusted by a mass flow controller before being heated by the electrical air heater to 540°C. Concentrated radiation provided by the HFSS heats the air inside the volumetric receiver before being expanded in the choked nozzle. Before conducting the experiments the receiver was covered with approximately 15 cm of high temperature insulation.
To calculate the receiver performance indicators pressure and temperature measurement have been taken at different positions. In Figure 2.13 can be seen connections for the pressure transducers (blue tubes), K-type thermocouples (green connectors), 4 pyrometers mounted in a 2-axis linear unit in front of a water cooled shield. The pyrometers are used to measure the receiver glass window temperature as well as the volumetric absorber temperature during operation. For the measurement, the sensors which are protected by the shield are moved quickly in front of the receiver with the lamps still operating while the air mass flow is turned off.
Tests have been performed reaching a maximum air outlet temperature of up to 820°C. At the nominal operating point the air temperature inside the receiver was increased from 540°C to 800°C with a conversion efficiency of approximately 80%.
Analysing the test it can be concluded that the volumetric receiver concept is a suitable candidate for the integration in the OMSoP system with a measured efficiency of 80% and the potential to reach even higher values.
The cavity receiver was coupled to a MGT supplied by CP in HFSS KTH lab in order to investigate the dynamic behaviour of the coupled system as well as to obtain improved measurement results of the cavity receiver performance.
The MGT provided by CP was coupled to the receiver mounted in the KTH HFSS. The MGT has a nominal electrical power output of 5kWe, a nominal speed of 160,000 rpm, requires a thermal input of 31.25 kW, and operates at a compressor pressure of 3 bar (absolute), with a nominal mass flow of 100 g /s with a turbine inlet temperature of 800°C.
Figure 2.14 shows the layout of the integrated MGT and receiver. Air is compressed in the compressor to 3 bar (absolute) before entering the heat exchanger (HEX) where it is heated to 500°C. Then the concentrated radiation heats the air within the solar receiver to 800°C or higher before being expanded in the turbine. In a next step, the air passes through the combustion chamber where additional heat is supplied to the system if necessary. Finally, the air passes through the heat exchanger and is then expelled at the exhaust.
The physical integration of the receiver and the MGT required a couple of meters of piping as shown Figure 2.15 Figure 2.16 and Figure 2.17.
Since the prototype receiver is originally designed for different radiative boundary conditions, the HFSS can only provide 30% - 42% of the radiative power required. Therefore, two methods have been applied to achieve the required absorber temperature during the testing: increasing the receiver inlet temperature and reducing the mass flow.
During the experimental campaign, multiple cases with different start boundary conditions were investigated and six representative ones are presented here. To avoid the quenching effect caused by the impinging jets, the MGT and the HFSS were turned off at the same time during the cooling down operations. Figure 2.18 shows the experimental results in terms of the absorber temperatures, receiver inlet temperature, outlet temperature, air mass flow and radiative power from the HFSS.
During most of the time the MGT ran at a low speed and the mass flow was controlled between 22 g/s and 26 g/s to achieve the required absorber surface temperatures and receiver outlet temperature. It was found that the receiver can produce outlet air of up to 810.5°C with absorber peak temperatures in the region of 960°C which match closely with simulations of the to the receiver performance in the OMSoP solar dish. Therefore, it can be assumed that the energy losses and cooling down rates are representative when compared to the solar dish system. During start-up it was found that temperature differences in the absorber cavity reached values of up to 100°C when the receiver is heated from ambient temperature. To avoid structural problems, it was concluded in the future real applications, this ’cold start’ should be avoided.
Additionally, to avoid thermal shock and to guarantee lifetime the temperature changing rate of the absorber cavity is important. It can be seen in Figure 2.19 that the absorber temperature and the receiver outlet temperature change quickly during the start-up and turn off.
In cases 2, 4, 5 and 6 the absorber temperatures have reached quasi steady states at similar levels compared to the receiver design temperature condition. Therefore, the energy losses of the receiver can be used for predicting the receiver total efficiency. Figure 2.20 shows the details of the energy losses for these four cases. It can be seen that the heat losses are significantly affected by the mass flow (see the curves of case 4, 5 and 6) as higher mass flows help to take up more heat by offering a higher velocity and Nusselt number. Material temperatures are another factor that affect the energy losses due to increased radiation and convection heat loss (see the curves of cases 2 and 6).
Considering the data stability, the average energy loss value of case 2 is used for predicting the receiver efficiency when it works under its designed boundary conditions. With the measured heat losses, the efficiency of the cavity receiver can be estimated to be around 73.3% for a solar that provides 35.3 kW in an aperture of 200 mm.
Analysing the integrated MGT-receiver test it can be concluded that the impinging receiver is suitable for the integration in the OMSoP solar dish system. It was shown that the impinging receiver can reach air outlet temperatures of 810°C at an absorber temperature of 960°C. The absorber cavity temperature uniformity is good with exception of the ‘cold start-up’ process. The total energy loss of the receiver is around 9.15 kW, and the efficiency of the receiver can be estimated to reach 73.7% at its design boundary conditions.
The micro gas turbine
At the start of the project, it was postulated that an existing micro-gas turbine (MGT) developed by CP will be adapted for the CSP system. This 7 kWe MGT was originally developed for external firing for combined heat and power applications allowing for fuel flexibility. The turbomachinery components are based on modified turbocharger components and the shaft bearing arrangement are of the cantilevered type where the turbomachinery is over-hung from one end with the high-speed generator placed between two bearings.
Initial studies on the system operation and rotor dynamics showed that this system will have a critical vibration mode close to the operating range. In a typical operation, the MGT will be accelerated fast past that mode and operated above the critical speed. With the variable solar irradiation, it was deemed that it is possible that the system will be frequently passing past that mode during operation which may lead to high vibrations and failure. Alternative arrangements and designs were considered and it was decided to design a micro gas turbine operating sub-critical, will all vibrations mode above the operating range. The final arrangement is shown in Figure 3.1 with the generator placed in between the turbine and compressor. A new turbine and compressor design were completed and turbine materials were selected to comply with the subcritical operation. All the necessary electrical and electronic equipment were designed and built including the control system. The main S & T achievements are:
A new turbine design was completed which was optimised for high efficiency and light weigh. This we achieved using an in-house design model which followed by design improvements informed by computational fluid dynamic, structural dynamics and thermal analyses. These aim at meeting the design requirements while ensuring the elimination of host spots and high stress concentration points. The turbine material chosen was titanium aluminide (TiAL). Figure 3.2 shows views and dimensions of the final turbine design while figure 3.3 shows pictures of two manufactured turbines, the left-hand side one is an Inconel turbine manufactured using an additive manufacturing method while the right-hand side is machined from TiAL. Tests on the turbine were conducted at the City University micro gas turbine rig built during the project to verify the predicted behaviour. Figure 3.4 shows a comparison of experimental data and predictions with good agreement.
A set of electrical and electronic equipment were designed and sourced/built that were necessary for laboratory testing as well final integration with the solar dish system. These include a rectifier, an electrical drive and a sign wave filter that would allow motoring mode to higher speeds than in conventional drives to provide enough air protecting the receiver during start up model. The turbine test rig and electrical equipment are shown in figures 3.5 and 3.6.
Design and testing of the receiver interface and recuperator including pipe network and heat exchanger was completed as shown in Figure 3.7. CFD tools were used to provide guidance to the design process. The results were used by ENEA to integrate the system for the demonstration phase. Flexible pipes were used to connect different components, i.e. compressor, turbine, receiver and recuperator. This is mainly to accommodate the thermal expansion in the structure.
A novel solar operated micro gas turbine control system was developed, programmed and used for the demonstration activity. The preliminary control strategy, limiters and control sequences have been identified by CP. A dynamic simulation of the system was performed which showed the effectiveness of the proposed closed-loop control strategy. The proposed control logic facilitates the electrical conversion system to control the operation of the MGT. Further work was done by City, ENEA and CP to enhance the control sequence and the electrical scheme of the system which is the main mechanism to control the system. The hardware of the controller was programmed at City and tested later at ENEA in the demonstration plant.
The Micro Gas Turbine has been designed by City and manufactured partially at City workshop. The manufacturing of most of the components have been outsourced to different firms. The assembled City-MGT was prepared for the demonstration test of at EAEA site in Casaccia Centre.
A thermodynamic model of the MGT-CSP cycle was developed by City team and integrated in an optimisation framework to optimise the thermos-economic of the systems. An innovative operation strategy was developed and studied to improve the performance of solar only MGT-CSP. Figure 3.8 shows an example of the optimisation results for a MGT-CSP system to maximise the annual generated electricity while maintaining the cost of produced electricity on its lower level.
Comprehensive rotordynamic study was carried out by the City team to optimise the dynamic stability of the MGT. three different mechanical arrangements were compared for the MGT design. The study included using of alternative material for the turbine wheel to improve the dynamic stability. Table 3.1 summarises the pros and cons of different mechanical arrangements adopted for the MGT.
The turbomachinery components were designed and optimised for solar applications where due to the variation in DNI the off-design performance of the compressor and turbine become more critical. The CFD analysis of the compressor was integrated with an optimiser to enhance the performance of the system over a wide range of solar irradiation. Figure 3.9 shows the positive effect of compressor optimisation for the part-load operation on the overall cycle efficiency.
A detailed flow study was carried out at City to understand the effect of fine-scale features on a radial turbine on improvement of its aerodynamic efficiency. A test rig was setup and a series of experiments were carried out which showed the positive effect of the proposed features in the efficiency of small radial turbines. Figure 3.10 shows the effectiveness of the proposed features on the radial turbine efficiency.
System integration and testing
Based on the outcome of the experimental characterization of the components, the limited optical quality of the dish, and the results of the system simulation, the design of the demo system was iteratively revised, both in terms of component specifications and integrated system configuration. Regarding the component specifications, the receiver was resized and the system parameters were revised to keep unchanged the target power of about 5 kWe even in presence of an unexpected low optical performance of the dish. About the system configuration, different aspects were considered, ranging from mechanical to electric and electronic integration. For the mechanical aspect, the focus was on the design of a supporting structure which could meet the specification of every single component, ensure the mechanical stability of the system, and guarantee a free thermal expansion in every different inclination angle (Fig. 4.1).
Furthermore, a big effort was spent by ENEA, with the contribution of City and CP, on the design of the integrated electrical schemes and equipment. This aspect was initially underestimated and caused a delay on the completion and realization of the demo plant. Several electrical schemes were conceived mainly to individuate the best way to connect the electrical output of the MGT and to design the Power Electronic device in charge of speeding-up the Micro Gas Turbine and to transform its electric output into a usable electric flow. This work played a significant role in the project development since the different choices impacted on the controller hardware design and realization. The final solution involved the adoption of 4 individual components, connected according to Fig. 4.2 to realise a new conversion system, not commercially available.
The main system equipment (receiver, MGT, diffuser, and recuperator) were delivered to ENEA in winter 2017 while auxiliaries, oil pump, tubes and connectors were available in the first months of 2017. The process of physical assembly in lab took about four months, ending in April 2017, since several adjustments and modifications were conceived on-going leading to a system mechanical assembly more functional than the initial design. In particular the supporting structure was completely revised once started the process of assembly, to guarantee at the same time the mechanical stability of the system and the free thermal expansion of the hot components (mainly the receiver). Furthermore, all the connections between the components were studied to ensure the assembly and disassembly operation in case of damage of some components. The supporting structure was realized with 4 mm thick stainless steel L profiles and each component was anchored taking into account different constraints.
Initially the MGT was meant to be sustained directly by the recuperator but during the assembly process it became evident that the LP inlet flange of the recuperator could not bear the MGT weight, since the casing of the recuperator was significantly deformed once connected the MGT. Therefore, a dedicated system was developed to hold the MGT/HSG group to the structure.
The final system configuration is represented in Fig. 4.3 where the insulation of the receiver and the recuperator are also documented.
The entire system was installed inside a metallic container, thermally divided in two sections, the hot and cold areas to protect the compressor, the HSG, the oil pump and the electrical equipment from the heat absorbed by the receiver.
The installation of the equipment on the dish arm was performed in June, to allow the instrumentation, calibration and preliminary testing of the whole system in lab. Initially the receiver/MGT group was lifted and anchored to the dish arm (Fig.4.5 a). Then protective panels around the receiver/MGT group were installed to create a protective envelope from environmental agents (Fig. 4.5 b).
After the installation of the MGT group, the experimental procedure for aligning the receiver window with the dish focus was replicated by ENEA since the weight of the System Container altered the previous centring.
From the electrical/electronic point of view, the integration of the receiver/MGT group with the dish required the connection of the DCU (Dish Control Unit) with the MGT controller (LabJack unit) and the installation of four electrical boxes outdoor, housing the system power supply, the Drive, the Rectifier, the Sinus Filter, the Data Acquisition System and the Controller hardware. The cabling of all these units was a little troublesome due to the different components into which the Converter Unit is subdivided. This fact led to long connections of the power and signal cables, with additional pathways for many types of electrical noise either from e.m. radiation and ground current paths.
For the demonstrative plant, which is aimed at verifying the technical feasibility of the integration between the solar dish and the MGTs, a sophisticated Data Acquisition System (DAS) was realized by ENEA, in collaboration with City and KTH), to deeply understand and evaluate the performance of this innovative technology. The DAS conceptual scheme is represented in Fig. 4.7 where the presence of two separate systems (DAS1 and DAS2) is put in evidence: DAS1 is aimed at acquiring the signals from the environmental sensors, while DAS2 is destined to register the process parameters. Furthermore, a third parallel system of acquisition was developed to directly send the value of the most important and critical process parameters to the controller (Labjack unit). The installation of the instrumentation took three months, starting from April to June. More than 50 process sensors were purchased by ENEA and City and installed on the demo plant, partly on the receiver/MGT group, partly on the electrical equipment, as represented on the scheme in Fig. 4.8 where the parameters acquired for control purpose are highlighted in red, while the signals destined to registration and experimental analysis are highlighted in green.
The calibration of the instrumentation was carried out contemporaneously with the sensors installation. Each sensor measurement chain was checked to verify the transmission of the right physical variable value. Even if the calibration procedure was completed in lab before June, once the System Container was installed on the dish and the demo plant operated, a further revision was necessary due to the presence of high and low frequencies noises on the signals, mainly generated by the inverter of the dish and the Drive. In particular the voltage signals coming from the Controller Remote Unit (Labjack) were significantly affected by the noise. Corrective actions on the grounding of the equipment were taken and RC filters were successfully installed on the Labjack module.
The testing on the demo plant at the ENEA Casaccia site started in the early 2016 to verify the resistance of the solar shield, a panel to be positioned in front of the OMSoP System Container to protect mechanical and electrical devices from the solar reflected radiation (Fig. 4.9). That component was initially foreseen but became a critical issue when, from the experimental flux distribution characterization conducted in the years 2015-1016, it was evident that the high concentrated solar radiation reflected by the dish only partially was intercepted by the receiver window, and a large part was spread around the focus.
In a second step, after the completion of the system assembly and instrumentation, a first series of tests was performed in lab to verify the correct operation of the sensors installed and to check the ability of the Drive to control the MGT speed rate. The speed rate of the MGT was increased progressively during the tests, from 40000 rpm (Fig. 4.10) up to 75000 rpm (Fig. 4.11) to stress the system gradually, to acquire awareness of each single aspect related to the system functioning. Some oil leakages were detected during the operation of the MGT, particularly at low speed levels. Corrective actions were taken by City and ENEA to fix the problem. Anyway, based on the experience gained on the management of the oil circuit, it was evident that an oil-free design of the MGT must be developed as a future improvement of the technology. From the experimental campaign it was also confirmed that vibrations get lower (Figure 4.11) when increasing the speed rate over 70 krpm: therefore it was concluded that during the start-up procedure of the demo plant the speed rate has to be brought very quickly over the 70 krpm threshold to reduce the mechanical stress on the MGT. Many other information and data were collected to guide the operation of the system when tracking the sun.
Due to the late installation of the MGT/receiver group on the dish arm, mainly caused by the delayed definition of the electrical schemes and the delayed delivery of the electrical equipment (June), control hardware and software (beginning of July), the on-sun testing activity was limited to the month of July. ENEA, City and CP collaborated in the definition of the operational sequence, both in automatic and manual modes. Initially the system was operated in off-sun position, replicating the same tests as the ones performed in lab. The focus was put on the evident signal interference occurring during the dish manual operation on the parameters acquired by the Labjack module: the range of variability was relevant, even in cold test conditions, and some parameters presented evident off-sets values. As a corrective action, a module of 14 RC filters was connected to the signals entering the Labjack unit to protect them from the high frequency noise coming from the inverter of the dish, the inverter of the air fan and the Drive.
As a first test, the dish was put on automatic tracking and the MGT was motored up to 70 krpm to verify the ability of the control software to command the single components, included the dish. Nevertheless, an off-set of 20° on the dish azimuth position was set to prevent the heating of the system and perform a test out of focus. The MGT speed rate analogue signal acquired by the control unit (Labjack) still presented a significant variation from the set value, due to an improper grounding of the Drive, which was subsequently rectified. From the subsequent “cold” tests it was confirmed that the critical MGT speed range for the vibrations is 50-60 krpm (Figure 4.12).
Once verified the full operability of the different system apparatus mechanical/electrical/control), the experimental campaign on-sun was started with the contribution of City.
The testing procedure was structured in two main steps:
an initial tracking phase with an off-set of 20° in the dish azimuth to allow the system to reach high speed rate in cold conditions and to check the functionality of each equipment
a full tracking phase, with the receiver window aligned with the solar spot.
In the first tests, in a cautionary approach, the receiver was put intermittently on focus to allow a gradual heating of the receiver’s walls and absorber. Some cloudy days helped in limiting the heating ramp of the receiver (Fig. 4.14) where the evident difference between the TIT and the receiver outlet temperature (RECV-T2) can be attributed to the thermal inertia of the flanges and the piping connecting the receiver and the turbine.
In Figure 4.15a and Figure 4.15b the temperature courses and the pressure drops detected along the recuperator are represented, respectively. Obviously, the pressure drops tend to get higher while increasing the TIT.
The pressure levels detected along the circuit are represented in Figure 4.16: the compressor outlet pressure is within the range 1.00-1.47 bar in the MGT speed range 0-80 krpm. The vibration levels were quite stable and well below the limit of 7 mm/s; only isolated peaks were detected in correspondence of abrupt variations of the MGT speed rate. Therefore, as a recommendation, the MGT speed rate has to be gradually modulated during the start-up operation.
Each test was stopped once the turbine bearing temperature approached the value of 100°C. In this regard, the turbine bearing valve in the oil distributor was fully opened to increase the oil flow rate and consequently the cooling effect on the turbine bearing. The incremented oil flow led to a reduction of the turbine bearing temperature with respect to the compressor bearing in cold conditions (up to 90 krpm). Differently from the tests performed in lab, the stator temperature did not increase sharply, most probably thanks to the adoption of a finned heat exchanger in the lubrication oil circuit. Other on-sun tests were performed with the aim of registering the system parameters (both process and electrical parameters) in different environmental conditions and identify possible criticalities to be overcome (Fig. 4.17 Fig. 4.18).
Some unexpected difficulties were encountered, such as a relevant oil leakage occurring during an on-sun test: From an in-depth examination of the oil circuit it was verified that the cap gasket of the oil tank was deteriorated, leading to copious oil leakages when the dish inclination was higher than 45° from the horizontal axis. The gasket was replaced with a new one and the problem was fixed.
During the experimental campaign, it was noticed that mechanical power was produced even at low TIT levels (100-300 °C): in correspondence of a MGT speed rate of 65-70 krpm and a TIT range of 100-250°C, the power produced was about 500 W. (see Fig. 4.19). In correspondence of 80 krpm the maximum power produced in the TIT range 100-270°C was 900 W (Figure 4.19). At lower speed rate levels the production of mechanical power was less evident, even if not negligible.
Based on a linear extrapolation of the 80 krpm curve, the mechanical power that can be produced by the Turbine of the MGT in correspondence of 800 °C is expected to be 4.5 kW. Unfortunately, it was not possible to experimentally verify the validity of such prediction and to characterize the system behaviour at TIT levels higher than 300°C (and receiver temperatures higher than 700°C), due to the unexpected overheating of the turbine bearing temperature in correspondence of TIT higher than 300°C. Indeed, from the experimental campaign on the demo plant it clearly emerged that the turbine bearing temperature increases very steeply once the system starts to heat up, following the TIT course. As an example, in figure 4.20 the temperature course of the turbine bearing (BER_T2) detected during an on-sun test is represented, along with the TIT behaviour. The heating ramp of the TIT was quite steep in this test, since the associated DNI was above 800 W/m2 (Fig. 4.21). After some minutes the dish was put out of focus to avoid the turbine bearing overheating. The bearings adopted in the present MGT design cannot stand temperatures higher than 120°C. Some corrective measures were undertaken to efficiently remove the heat from the bearings, such as increasing the oil lubrication flowrate and pressure, but no evident results were obtained. Indeed, even if the oil lubrication inlet temperature was kept substantially constant during the tests, thanks to the adoption of a finned heat exchanger, the cooling effect in the turbine bearings was not sufficient. Due to time constraints, it was not possible to rectify this issue within the project duration to increase TIT levels.
However, the operability of an innovative system which can generate electrical power from solar irradiation was proven through the experimental campaign (Fig. 4.22). The management of the operating procedures through the control software was successful and the mechanical and thermal resistance of the components and the whole assembly was verified up to a speed of 90 krpm, a TIT of 300 °C and receiver temperatures up to 700 °C. Empirical system parameters, such as critical speed rates (50-60 krpm), duration of the defocusing procedure (15 seconds), thermal inertia of the components, lubrication pressure (2,5-3 bar), were identified. Based on the results of the experimental campaign, an in-depth analysis of the system performance, both in terms of stationary and transient behaviour, will be performed in the future. Furthermore, with the knowledge gained through the operation of the demo plant, several recommendations in view of a future improved design were identified.
System optimisation
The steady state plant simulator for design purposes has been developed using the models and approaches presented in the previous periods. The plant simulator contains the various steady state modules associated to components as shown in the Figure 5.1.
The Model of the environment establishes the air quality (dry air composition and relative humidity), temperature and pressure. Moreover, the sun DNI is defined.
It is considered as a parabolic shaped solid dish as shown in Figure 5.2 it is characterized by sizes: diameter, reflecting surface, by optical performance: efficiency, maximal temperature of the spot related to the DNI value and shape of the spot. For this application, a perfect tracking system is considered, thus, the tracking error is set to zero.
The receiver is the device that transfers the radiant heat power concentrated at its opening to the Working Fluid entering it, and to the environment as lost heat power. The component model is sketched together with an example of cavity receiver in Figure 5.3.
The MGT module contains the MG selection and sizing module that had already been completed. It is based on the block diagram of Figure 5.4 that allows to establish the sizes of turbomachinery and the related maps according to the impeller technology database, an example of such a DB is shown in Figure 5.4. The engine related equation set represents a correlation among performance and sizes of subcomponents. The internal quantities are also calculated to be used for other purposes. Various complex MT arrangement can be selected.
Figure 5.5 shows the relationship between efficiency and electric current for the whole electric subsystem. The section concerning the auxiliaries leads to establish the power absorbed by the auxiliaries on empirical basis.
All these module equation sets are solved by a modular approach. The degrees of freedom are optimized according to the minimization of a suitable objective function. The scheme of the process is shown in Figure 5.6. It allows the cycle calculation using the calculated efficiencies of the components by an iterative procedure. After the components have been sized, their off-design behaviour is established. Using all the off-design modules the matching problem among them is solved. Results represent a steady state running point of the plant, the boundary conditions being assumed.
For the OMSoP demonstration plant, the Figure 5.7 shows the part load behaviour versus the DNI. The left-hand figure gives the produced power, the CSP efficiency, the MGT rotational speed and the mass flow rate. The right-hand figure gives the electric efficiency, the receiver efficiency and the product of the MGT compressor, expander and mechanical efficiencies.
A comparison between the Stirling CSP performance and the MGT-CSP ones has been done. Stirling CSP efficiency is given versus electric power in Figure 5.8. The above steady state simulator has allowed the evaluation of the various CSPs equipped with the complex MGT arrangements. Results are shown in Figure 5.9 it shows that for both Stirling and MGT, the higher the power is, higher the efficiency is. Increasing the MGT arrangement complexity, especially for the higher power (35kW), the MGT-CSPs have better expected performance than the Stirling CSPs.
To improve the overall CSP plan performance reducing costs, some MT-CSP combinations with ORC have been analysed.
First, a complete investigation focused on ORC cycle net efficiencies has been carried out checking in the real applications. Figure 5.10 shows the ORC net efficiency versus the temperature of the heat source. Various combinations of the ORCs and the CSP-MGTs have been studied taking as reference the OMSoP demonstrator.
A first combination is shown in Figure 5.11 that shows an ORC cycle bottomed to the MGT. Since the MGT exhaust can have a temperature that ranges from 250°C to 300°C, an average ORC efficiency of 13% can be expected. Thus, the ORC turbine can produce some 3kW. The demonstrator global power could become about 9kW and the overall efficiency can increase of about 44%.
A second proposed combination scheme is depicted in Figure 5.12. It is for a combination between the MGT and an ORC that takes the heat power from the MGT exhaust and from the focal spot by a dedicated tubular receiver being irradiated by the part of the spot, whose irradiance levels lead to temperatures lower than that required by the MGT receiver. This heat power can be stored at medium temperature suitable for the achievement of good ORC efficiencies. Such a stored energy can be used to compensate the sun irradiance fluctuations that influence the TIT and the MGT power output.
Another proposed arrangement has been analysed for 24h working by mean of a medium temperature concentrator and thus lower cost per kW of heat power has been foreseen. The combination of the two cycles can lead to operate the plant 24h per day with an average efficiency intermediate between the combined MGT – OR simply combined cycle and the ORC one.
Market and Cost analysis
The economic performance of the OMSoP system can be assessed through several variables, the most relevant of which are the Installed Cost of the system (€/kWe) and the Levelised Cost of the Electricity (LCoE) produced with it. These items have been calculated independently with the aims to estimate both contributions as accurately as possible and to assess the driver of system economics.
The economic analysis of any technology that is still at a moderate TRL level is inherently uncertain and based on numerous estimates and forecasts which might become or not true. Therefore, it is mandatory to account for uncertainty and to assess how much the final results might be impacted by deviations in the input data set. This was done in two different ways, through the direct parameterisation of the results (i.e. by changing the input variables in a certain range) and with a simple statistical analysis based on the Monte Carlo approach. The latter is used to implement uncertainty in the market analysis whilst the former is employed in the direct estimation of costs.
The analysis described in the previous paragraphs was carried out for two business cases, namely the generation of electric power only (Business Case #0) and the simultaneous generation of heat and power (Business Case #1). Each business case is then divided into two possible operational modes: solar-only, where the system is run with solar energy only, and hybrid, where fossil fuel is burnt occasionally to make up for the lack of solar energy during cloudy periods of time or at night.
From a technical standpoint, it is also considered that the OMSoP technology can be implemented with moderate specifications and more high-performing cycle parameters. Two figures of merit are used to identify the standard and advanced systems: turbine inlet temperature (TIT) and recuperator effectiveness. The resulting combinations are shown in Table 6.1
The four cases shown in Table 6.1 are carried out through the entire analysis but, for the sake of clarity, most of the results are referred to just two of them. Indeed, the grey-shaded cells in the table stand for the reference or base-case system (800ºC and 85%) and the advanced system (900ºC and 90%). These are also referred to as Level I and Level II systems respectively.
Technology Levels I & II are initially implemented in a simple recuperative cycle with one compressor and turbine, a shown schematically in Figure 6.1. Nevertheless, there might be a value in adopting more complex layouts which can be implemented with essentially the same technology at the component level (Table 6.1). These complex systems are:
Intercooled recuperative system (ICR) whereby the compression process is split into two stages with an intercooling heat exchanger in between. The system still makes use of a recuperative heat exchanger.
Intercooled and reheat recuperative system (ICRR). In this case, the expansion process is also split in two expanders in series in between of which a second solar receiver is located.
There are three main items regarding the final cost of the electricity generated with OMSoP that must be supported by the end-user. These are presented below in chronological order as they have been developed in the project:
Installed cost of the technology. This is expressed in €/kWe. It accounts for the capital cost of the main equipment plus transportation and installation, including import/export duties.
Financial costs of the project, brought about by the operating costs and the finance-related costs like banking and debt pay-off. This contribution depends mainly on the boundary conditions under which the system is operated.
Yield. This is the amount of electricity generated which depends on the thermodynamic performance of the system and on the available solar resource at the system’s location.
A sketch outlining how the total installed cost is calculated is presented in Figure 6.2. It shows how the cost of the OMSoP system is broken down its main constituents to calculate the so-called Purchase Equipment Cost, which can be regarded in this context as the capital cost of the system ex-works (at the factory site). As evidenced, this cost depends on the layout and specifications of the system (see Table 6.1) and the manufacturing volume.
Component costs are calculated based on two main influential factors, one technical feature representing the size of the component and another functional dependence on the manufacturing volume as evidenced in Figure 6.2. The figure of merit for size is, of course, different for each component: mass flow rate for the gas turbine, diameter for the size, thermal output for the solar receiver and electric output for the Balance of Plant (BOP) components. Based on this approach, a complete literature review is performed to gather as much information about costs related to the OMSoP components. This means not only experience with dish and MGT integrated systems, like the work carried out by NASA in the 1980s, but also other systems using individual OMSoP components like dish-Stirling systems (cost of the dish), micro gas turbines and even solar receivers for process heat. The information obtained in this literature review is then completed with information provided by the OMSoP consortium members specifically.
The resulting database is then integrated and cost functions are developed for each component with the previously mentioned sensitivity to size and production volume. Some of these functions are shown in Figure 6.3 where the impact of the technology level is easily observed.
Further to this Purchase Equipment Cost, there are other cost items that must be added to the system: transportation, import fees and installation being the most relevant. These additional costs are driven by the location of the final system and they can largely influence the final installed cost as shown in Table 6.2.
The impact of location and technology/layout on the installed cost of the system is presented in Figure 6.4 for three locations: South Africa, Morocco and the East bank of China. The following nomenclature is used:
BC# stands for the business case: the code is 0 for electric power generation and 1 for combined heat and power.
SR / ICR / ICRR stands for the engine layout: simple recuperative (SR), intercooled recuperative (ICR) and intercooled and reheated recuperative (ICRR).
I / II stands for the technology level: Level I is for 800ºC turbine inlet temperature and 85% recuperator effectiveness and Level II for 900ºC and 90%.
The impact of location in figure 6.4 is evident. More favourable locations, i.e. locations with a more abundant solar resource, yield lower Installed Cost due to a smaller dish diameter or, reversely, a larger output for the same dish size. The final effect on cost can be as large as 10% if feasible locations are considered. Whilst this was to be expected, the impacts of technology level and engine layout are not so evident. Indeed, it is observed that there is a drastic drop in installed costs when higher components specifications are selected (upgrading from Level I to Level II). This is marked as (1) in the chart. On the contrary, the potential benefits coming from a more complex layout are not as large as initially expected. The transition from a simple recuperative system to a more complex engine using intercooling and reheat seem to have a much weaker impact on cost as deduced from the almost constant cost marked with (2) in Figure 6.4.
The information in the mentioned cost analysis is then integrated with the performance model developed to carry out annual simulations of the system in different locations. This is usual practice in project appraisal since the utilisation of annually averaged values of efficiency and insolation yield inaccurate results. The techno-economic performance of OMSoP in the selected locations is presented in Table 6.3 below with the same nomenclature used before.
There are several features in Table 3 that are worth noting:
The capacity factor (fraction of hours in a year that the system is in operation, referred to equivalent full load operation) and efficiency are strongly dependent upon location.
There is a large gain in efficiency when the system is upgraded from Level I to Level II technology.
When hybrid operation is considered (BC #0 – H), there is the potential to achieve 50% capacity factor with 90% solar share (minimum contribution from fossil fuels).
When Combined Heat and Power is considered, the electricity to heat ratio is raised with respect to other competing technologies like reciprocating engines. In particular, a 50% split is affordable with the standard Level II technology.
The same information is now presented in Table 4 for the intercooled configurations of the engine. The main observations when comparing the information with that in Table 6.3 are:
Annual efficiencies close to 20% are possible.
The electricity to heat ratio in CHP applications increases to over 70% due to the much higher solar-to-electricity efficiency.
The information in Table 3 and Table 4 translates into a Levelised Cost of Electricity as stated in the first paragraph of this section. This information is presented in Figure 6.5 where it is credited that the cost of electricity generated by OMSoP can be substantially lower than 10 c€/kWh in a good location and 10 and 15 c€/kWh in a mild one.
The information in Figure 6.5 becomes even more interesting when compared to the existing technologies in the market that are competing for the same applications and end-users. Such comparison is illustrated in Figure 6.6 showing that that:
OMSoP is more cost-competitive than dish-Stirling systems when performing the comparison for the same location and set of boundary conditions.
With respect to photovoltaic technology:
OMSoP exhibits the same performance as photovoltaic panels in a good location with moderately high direct normal irradiance (cases of South Africa and Morocco).
Photovoltaic technology is more cost-competitive in locations where the rated DNI is lower. This is due to this technology being able to exploit diffuse radiation.
These bullet points confirm that the initial objective of the OMSoP project to develop a Concentrated Solar Power technology to produce Solar Thermal Electricity that is cost-competitive against the existing technologies at the small-scale is accomplished.
This work was aimed at the screening and identification of potential markets that could be of primary interest for the commercialisation of the OMSoP technology. Such analysis is based on the combination of certain drivers influencing the likeliness that an area of the world becomes interested in the technology. These drivers are:
Availability of the solar resource (F_I), in particular in the form of Direct Normal Irradiance. Only areas with high DNI are expected to become interested in solar thermal technology.
Size of the eventual market (F_D), i.e. number of systems or total installed power that could be deployed in a particular country. This is directly related to the demand for electricity and, indirectly, to the population of the area.
Characteristics of the national grid (F_G). Countries where a fraction of the population does not have access to electricity are likely to become more interested in OMSoP.
Renewable energy policy (F_P). Countries with legal initiatives to foster renewable energies are more interesting as initial markets for OMSoP.
Financial risk of the country (F_F). As in any economic initiative, assuring the financial and market conditions of a country is fundamental if a new product is to be commercialised and capital is to be invested.
These factors are combined according to certain weights (which prioritise ones over the others depending on the cases considered) yielding the so called Index of Market Potential IMP.
A database with information compiled from scientific publications, national regulators, the World Bank database, Eurostats, the International Energy Agency and other institutions, to characterise these factors F_i as accurately as possible for a list of countries representative of all the areas of interest in the world. This information is then implemented into an Application whereby a flexible environment enables changing the weighting factors of the IMP for different boundary conditions. The graphical user interface is shown in Figure 6.7 where the different menus to change the weighting factors are visible as so is the option to assess an individual, stand-alone system or an array of several units (so-called farm arrangement) to produce hundreds of kilowatts.
A thorough analysis was performed to set a realistic/accurate set of boundary conditions because of which trustworthy information can be obtained. A sample of the results is presented in Figure 6.8 where temperature maps of IMP are shown for the stand-alone and farm arrangements along with the particular values for a list of selected countries. There are several observations that are worthy of note:
The markets of primary interest differ from one arrangement to another, in particular at the top of the list. The least interesting countries seem to be more or less the same in both cases.
Market stability and energy policies seem to be equally important to availability of solar resource and market size when it comes to the IMP of individual units.
Large countries with uneven distribution of the solar resource tend to be heavily burdened by the least favourable areas. Therefore, it is advised that these countries be divided into internal regions where the same methodology can applied with an input dataset that is specific to the region.
With regards to the most interesting countries for the future commercialisation of OMSoP, the analysis carried out revealed the following:
Countries like South Africa, Chile, Australia, United States, Morocco, Spain and Mexico seem to have a large potential for the OMSoP technology.
Countries with sparsely distributed population and high solar resources like Algeria or Morocco are also of interest for the stand-alone OMSoP system.
Countries with sunny regions that are lightly populated or large distance between regions with high availability of solar energy and regions that are densely populated are not as interesting; such is the case of Brazil or Italy. For these, large CSP facilities using other technologies are probably more cost effective.
Power plants made up of large arrays of OMSoP systems might find market opportunities in countries with high electricity demand and more favourable market conditions.
Uncertainty analysis is a transversal task which, from a practical standpoint, is directly implemented into the previously described cost and market analyses. The impact of uncertainty is assessed from two different angles. On one hand, a deterministic quantification of uncertainty can be implemented by merely performing sensitivity analysis with respect to the main input data set. This is used mostly in the cost analysis where the economic database compiles information from very different sources for each cost item which, in turn, can potentially take values within a range. How wide or narrow the range is depends on the particular item whose cost is being estimated.
Figure 6.9 illustrates this. The information in the charts shows the raw data that is used to produce the individual cost functions that were shown in Figure 6.3. It is easily observed that the different sources of information provide specific costs that are, in some cases, largely dissimilar. Such is the case of the micro gas turbine for which the lines in the upper plot are quite distant from one another, yielding important deviations in specific cost. On the contrary, the costs gathered for the receiver seem to be in good agreement.
The data in Figure 6.9 are weighted averaged considering the time value of money and the natural development of the technology with time. Then, a sensitivity analysis is performed to assess how much the final results in Figure 6.4 are affected by changes in the cost functions presented in Figure 6.3. The sensitivity analysis shows that the impact on the results is one order of magnitude lower than the main figures of merit, thus confirming that the results presented in the cost analysis are trustworthy.
The second angle from which uncertainty is approached refers to the market analysis. Indeed, market analysis is inherently uncertain because it takes into account future actions that might or not happen. Even if the current scenario were described and quantified very accurately (financial conditions for the investment, current number of end-users, and current demand of electricity in a country...) the future behaviour of the market would still hold a large degree of uncertainty; numerous examples can be found in handbooks covering this topic. For this reason, a more complex approach to uncertainty management is needed.
To manage uncertainty in the market analysis, the Monte Carlo method is used. This method relies on the same deterministic principle as the approach used in the cost analysis (see previous paragraph) but adds a statistical or random feature to it. Thus, the input variables do not take discrete values but, on the contrary, they take values in a range with a probability function assigned. When the impact of uncertainty is assessed, the same application described before is run thousands of times, each one of which sets up the input variable set with samples from the cited probability functions. If the number of runs is high enough, one can be confident that the whole range of values has been tested, even those with a lower probability, and the impact on the results can be evaluated.
The practical implementation has been made with standard probability functions and with other functions, yielding somewhat similar results. This is observed in Figure 6.10 where detailed information about the market analysis for India is presented. The statistical nature of the assessment is evident in the upper row of charts where the input variable values covered in the Monte Carlo analysis is presented. As shown, the analysis is run with a very wide of input data. The bottom row presents the position occupied by the country in a ranked list according to the countries’ IMP. The impact of varying some of the input factors F_i is strong (the position in the rank list changes) whereas other factors have a very weak effect.
The information presented graphically in Figure 6.10 can also be presented numerically. It is very interesting to see that even when large uncertainty is added to the input dataset, the resulting list of countries ranked according to the Index of Market Potential remains quite unaltered with just minor changes one or maximum two positions up or down.
Potential Impact:
The OMSoP project has demonstrated the technical and economic feasibility of integrating a concentrated solar power parabolic solar dish system with a micro gas turbine to generate electricity. The electrical power output from a single unit with one parabolic would be in the region of few kW to about 25 kW. This power range for a single unit is suitable for small to medium scale domestic, industrial and public services applications. The units are modular and hence can be stacked to produce larger power outputs if needed, potentially in the region of Megawatts to tens of Megawatts covering a gap in the available solar thermal power technology which is only economic for much larger power outputs. In comparison to the popular photovoltaic systems, the OMSoP system can be transferred to a fully dispatchable power plant through hybridisation with fuel combustion or thermal energy storage, or possibly both in the same units to achieve reliable source of electricity. The most suitable locations in the first instance would be off-grid locations or those with low grid reliability. This however does not preclude the system from direct competition with other renewable energy systems connected to the gird as the techno-economic studies have shown.
At the end of the project, it can be claimed that the proposed OMSoP technology has achieved a technology readiness level between 6 and 7 where a prototype was tested in an environment very close to operational one.
The project has addressed three key impact areas through the dedicated technical work packages. These are:
1. System component development (Work Package 1), were system components were developed: receiver, concentrator and MGT to provide technical solutions to issues that have previously prevented the wide spread deployment of similar systems. For instance, a high efficiency micro gas turbine was developed that would lead to reduction in dish size which is the single most expensive part of the system.
2. System design and integration (Work Package 2), aimed at integrating the components developed in WP1 into a functioning system by system design and optimisation, also aimed at demonstrating the system which provides a proof of concept and feedback for component optimisation. The outcome of this activity has been extensive knowledge gained in the field of solar powered micro gas turbines in terms of sourcing of components, manufacture, integration, electrical scheme, control system and many other detailed aspects that would provide clear direction for future work towards commercialisation.
3. Techno-Economic analysis (Work Package 3), aimed at developing thermodynamic and mechanical models of the system and its components, as well as further insight into future deployment by studying concepts such as medium and long term storage hybridisation with other fuel and MGT power argumentation. Another objective has been to identify the potential markets for such systems worldwide just paving the way for targets for initial deployment.
The ultimate impact of OMSoP is to introduce a step change in the current paradigm of Concentrated Solar power technology which, if the current techno-economic scenario remains, will progressively be less competitive compared to PV and wind. The OMSoP concept, which aims to widen the scope of solar thermal technology from the current emphasis to centralized generation towards distributed generation targeting small and medium sized applications. This will tackle two major hurdles in the current solar power market. On one end, compared to centralized generation, it will reduce significantly the cost of the grid infrastructure particularly in countries with many small scale scattered communities. This will also reduce grid losses and allow for combined heat and power, combined cooling and power or multiple generation at the point of consumption hence increasing the overall efficiency of the system. On the other hand, and compared to PV used in distributed generation, the OMSoP system has the advantage of flexibility to be extended to hybrid systems with co-firing and/or thermal energy storage allowing for potential full dispatchability, compared to PV which needs either another system such as diesel engines or battery storage, the cost and environmental impact will be significantly lower.
The successful completion deployment of the demonstrated OMSoP technology thus could contribute to a new scenario of the Energy sector in several regions in Europe and many other countries with high solar irradiation wherein this type of energy technology will ensure dispatchability and cost competitiveness simultaneously. This has not taken place yet since PV and wind are cost-competitive but not dispatchable whilst state-of-the-art CSP is dispatchable but has relatively very high costs and only suitable for centralised power plants. Indeed, dispatchability sets CSP apart from other technologies with regards to increasing renewable energy penetration. This has been confirmed by a study carried out by IEA and IRENA (Renewable Energy and Jobs. Annual review 2017. IRENA) where it is claimed that the overall installed capacity in CSP will increase from the actual 4 GW, to 100 GW by 2030 and 300 GW by 2050, resulting in a yearly installed capacity of 10 GW in 2030 and 20 GW in 2050. The assessment of the cost of electricity from OMSoP has been shown to be competitive with PV and conventional power generation through the studies conducted in work package t3 of this project. This suggests that if the right interest by industry and investors is achieved through appropriate dissemination of the outcomes of the project, it is more than likely that the proposed system will make its way towards commercialisation.
The following list evidences how OMSoP has addressed the expected specific impacts put forward in the European Comission work programmes
• Reduce the technological risks for the next development stages: This project has demonstrated that the proposed technology is feasible from both rechincal and economic points of view. For the first time worldwide a micro gas turbine of the proposed power range was successfully tested operating with concentrated solar power using a single parabolic dish concentrator. All the gained knowledge on concentrator, receiver, micro gas turbine, electrical eclectronic and control componentes, materials and component manufacturing process has provided new knowledge and experidence that will reduce the technological risk for the future comercialisation activity of this technology.
• Significantly increased technology performance: the utilization of a micro gas turbine parabolic dish concentrated solar tower technology for distributed and small scale centeralise power has shown to yield competitive technologies in terms of cost and reliability. The gain is expected to be higher if the cost of land is taken into account where the proposed system requires about half of that needed by PV for the same power output.
• Reducing the life-cycle environmental impact: the deployment of OMSoP will contribute to the widespread deployment of CSP technology which is renewable and with negligible CO2 footprint. Compared to other solar power technologies, OMSoP will further reduce its life-cycle enviromental impact as a result of the smaller power block, therefore requiring less materials, and most importantly, lower land use due to the smaller solar field.
• Nurturing the development of the industrial capacity to produce components and systems and opening of new opportunities: Industrial partners will be actively involved in the development of OMSoP technology for future deployment. Further European industry is expected to join future commercialisation activity including those from Micro gas turbine and solar dish manufacturers.
• Contributing to the strengthening the European industrial technology base, thereby creating growth and jobs in Europe: the development of an innovative CSP technology will take Europe to the forefront of the renewable energy industry. Indeed, European industries have led CSP since the early 2000’s. In order to strengthen the position of the European industry and ensure competitiveness against the fast-growing Far East industry, the development of highly innovative concepts must become a priority in the agenda of European decision-makers. The unfortunate experience in the PV industry, where most of EU manufacturers closed down when challenged by the Chinese industry, must be born in mind in this respect. Certainly, the project presents a groundbreaking concepts successfully demonstrated and also innovative methodologies that would give an edge to the European industry.
• Reducing renewable energy technologies installation time and cost and/or operational costs, hence easing the deployment of renewable energy sources within the energy mix: the cost of electricity reduction will be achieved through the higher overall conversion effiency, when taking into account utilisaiton of the dwonstream heat resulting even more competitve technology.
• Increasing the reliability and lifetime while decreasing operation and maintenance costs, hence creating new business opportunities: the OMSoP technology will achieve these targets through the more compact and reliable micro turbine compated to Striling engines. The other markets that can take advantage of the OMSoP technology. Additionally, micro gas turbines can be used for many other applications such as CHP, auromotive range exteders and back up power systems resulting in mass production and reduction in unit price.
• Contributing to solving the global climate and energy challenges: the OMSoP consortium is very concerned to the energy-related environmental problems that will be faced by future generations. This is why the technology exhibits numerous features oriented to solving or at least mitigating these problems: (i) OMSoP aims at contributing to a renewable-energy dominated world through the worldwide deployment of Concentrated Solar Power, (ii) within the outcome of the project, a dedicated studies covered the environmental performance of the technology, including Life Cycle impact and Greenhouse Gas emission reduction; (iii) a key objective of the project has been the simultaneous optimization of the overall solar plant efficiency in order to allow an LCoE reduction with respect to other technologies, therefore making solar electricity affordable including the case for developing regions of the world (key aspect of the energy challenge since environmentally clean electricity is still generally more expensive than electricity from fossil fuels).
• Other impacts OMSoP would ultimately lead to a fully dispatchable Concentrated Solar Power technology that is cost-competitive against any renewable energy technology and, as such, it will contribute to the diversification of energy sources in Europe. Additionally, due to the hybrid operation capability, it can partly or entirely replace conventional power plants. It must be noted that not only can the OMSoP concept be used in Concentrated Solar Power applications, but it can also be employed with other renewable energy sources such as biomass in co-firing or thermal energy storage.
Impact on developing countries and regions:
The electricity distribution networks are under developed in many parts of the world. In these areas: power grid network structure of high-voltage substation are very weak and their supply is not reliable; there is a lack of sub-stations, which causes each station’s coverage area to be too large and main substations are regularly out of repair; supply area of each site is not clear and power lines can be in chaos and power losses are typically in addition to mismatch between high voltage main lines and branch lines leading to unreliable supply. It can be inferred that, electricity supply cannot be guaranteed in poor areas, that is, frequent power outages and chronic power outages are common; voltage is low and no power supply during peak times. There are several disadvantages for off-grid household PV system complemented by wind; firstly, power availability is climate dependent. Additionally, low power density which would occupy a large area for higher capacity power system leading to high initial cost estimated at 3~15 times higher than conventional power generation systems. Secondly, large-scale application of wind-solar complementary power system is limited by high cost of energy storage systems if around the clock power is to be provided. High cost of pumped-storage systems which are also cite specific, electrochemical energy storage like lead-acid battery storage and their relatively short lifetime makes them economically not viable. On the other hand, many of these areas have high solar energy resources, but these can be with no electricity supply or insufficient power supply. Commercialisation of the OMSoP technology in particular, when coupled to thermal energy storage can lead to transformations in such societies at different levels. The impact to local communities in the following is conceivable, sometimes interdependent in relation to providing sustainable development, reducing poverty, improving the quality of life and health provision as follows: (1) the provision of clean reliable and affordable source of electricity using an abundant and sustainable resource. (2) The provision of local jobs through employment in the manufacture of system components, assembly, installation, operation and maintenance in addition to satellite jobs in the supply chain and services. This would help to improve the welfare of the local populations and reduce poverty. (3) Freeing up populations time from fetching primitive energy resources or coping with unreliable energy, which would allow more time for education and training providing better jobs and higher pay. (4) Availability of electricity for services, in particular health, would improve health, raise life expectancy and improve quality of life for people and reduce the cost of health services to local governments freeing up resources for further development.
Dissemination
To maximise the impact of the OMSoP project, the team has embarked on a wide range of dissemination activities to explain the project and increase future interest from potential industries, investors as well as customers. The following dissemination activities were undertaken:
• Project website (https://etn.global/research-innovation/projects/omsop/): The project web site was made live at the start of the project and will remain active for several years after the end of the project. It contains information about the technology, its operating principles and advantages as well as a description of the project structure. All project public reports are also made available on the website of access of the public.
• A trifold leaflet distributed to tradeshows and exhibitions (e.g. ASME Expo Turbo (2014 Dusseldorf, Germany, 2015 Montreal, Canada), POWERGEN (2014 Cologne, Germany; 2015 Amsterdam Netherlands), International Gas Turbine Conference (Brussels, Belgium), ETN’s Annual General Meetings in 2014 and 2015, as well as at all ETN meetings in Brussels.
• A presentation of the OMSoP Project and results at ETN’s AGM in Paris (2014) and in Dublin (2015).
• A presentation of the OMSOP project at the National Research Centre for Gas Turbine and IGCC Technology of China, January 2015, Beijing, China
• OMSOP project slides included in ETN’s general power-point presentation disseminated at various meetings and events globally (2013-2017).
• OMSOP project information and links on ETN website, in ETN’s Brochure and ETN’s Members package.
• A bi-monthly electronic update sent to the consortium with the status updates, the project reports and meeting information.
• Two video describing how the OMSoP system works, the objectives and the result achieved in the project and the challenges overcome in the project by the consortium partners.
• A press release of the OMSoP project highlighting the objectives achieved.
•
• Publications of articles on the worldwide press (e.g. Gas-To-Power Journal, Gas Turbine World, etc.).
• ETN continuously disseminates the OMSoP project progress and results through the ETN network via its website, Twitter, LinkedIn, Quarterly Newsletter and at the Annual General Meeting.
List of Websites:
https://etn.global/research-innovation/projects/omsop/