Final Report Summary - PERFORMANCE PLUS (Tools for Enhanced Photovoltaic System Performance)
Executive Summary:
The Performance Plus project was structured in two main phases. In the first phase, a series of devices, models, methods and tools to assess and optimize the performance and lifetime of photovoltaic (PV) installations have been developed. In the second phase, the results have been validated and demonstrated thanks to a User Group composed of large portfolio owners and strong EPC and O&M contractors that gave access to 25 PV systems across Europe making a total of more than 50 MW of installed capacity. In a nutshell, the final results of the project are:
Tools and methods to help reducing the risk during design phase
A method to combine satellite derived irradiation and ground measured data called Smart Irradiation Service has been developed. Up to 30% improvement in accuracy has been achieved by applying this method. Moreover, an innovative multi-directional irradiance sensor called ESA, that measures both the direct and indirect solar radiation was designed and built within the project. ESA’s smaller size, easier installation and operation and up to 85% of cost reduction compared with an equivalent solar irradiation station are amongst the advantages of this innovative sensor.
The new indoor PV module and inverter test procedures developed by the consortium help to characterize better the PV system components for real operating conditions and to discover quality issues before installation. In addition, thanks to the fine-grained temporal and spatial models that are able to take into account very complex environmental conditions, the overall uncertainty during design phase is reduced.
Tools and methods to increase the performance during operation phase
The consortium has developed a method to detect and diagnose faults and degradation remotely using monitoring data. The application of this method called PV Health Scan can increase the performance up to 10% thanks to the early detection of issues. Furthermore, the new toolbox for on-site testing of PV modules and inverters developed by the consortium allows to check the components directly on-site without the need of sending them to a laboratory for testing.
Tools and methods to allow a better integration of PV into the power grid
The consortium developed new methods to forecast imminent irradiance changes thanks to a sky imager that extracts the cloud distribution and their motion as well as the resulting surface shadow and solar irradiance fields. In combination with the fine-grain simulation models developed within the project, the accuracy of PV energy yield nowcasting methods is significantly improved. Moreover, the application of a Model Predictive Control (MPC) framework that allows harnessing flexibility from any controllable load enables a better integration of PV plants into the power grid. As demonstrated by the consortium, the application of this method can save up to 30% on heating energy costs during winter in Belgium.
The Performance Plus consortium presented the final results of this three years project to an audience of industry and researchers during a closing event, held in Brussels on the 6th of October 2015. The presentations given during this closing event and all public results are available on the project website: www.perfplus.eu
Project Context and Objectives:
Beyond Improving Components: System Level Optimization
The PV industry is today at an interesting point in its history. To ensure a continuous decrease of the costs linked to electricity from PV, the prices of modules, inverters and balance of system (BOS) components have to be further decreased. This has to be accomplished, while performance, functionality, reliability and lifetime on the component and system level need to be increased. Needless to say, the industry has made large technological progress with PV cells, modules and inverters in terms of costs and reliability. However, from an integrated perspective, PV system performance emerges from, but is not limited to, the performance of the components. Therefore, in order to improve the performance of PV systems, we need to look at how we can improve the system as a whole.
Limitations during Design Phase: Predicted vs Real Performance
There are several PV performance modelling software packages available on the market, specially developed to predict the amount of energy that a PV system can produce. The predicted performance from these models can differ significantly between each other and from actual performance from real systems. One of the main goals of this work was to better understand these discrepancies between predicted and actual performance of PV systems.
The large amount of input parameters like irradiance, temperature, array orientation, module and inverter performance, user-defined values for additional losses such as soiling, mismatching, cabling, etc. have inherent uncertainties. These uncertainties have to be properly taken into account, as their correct quantification is essential for evaluating the financial risk of PV investments.
Improving Operation & Maintenance: Early and Reliable Detection of Operational Performance Issues
For PV professionals, the early detection and diagnosis of faults is of the utmost importance to obtain and maintain high energy yield of PV systems. Moreover, the timely remediation of faults not only restores the production promptly but also avoids the occurrence of additional component failures and leads to reduction of costs for operation and maintenance (O&M). However, this early and reliable detection of issues requires accurate models of the expected behaviour of the well-performing PV plant. In practice, tools are required to help PV plant owners and operators reducing the occurrence of component failures and thus reducing O&M costs. Ideally, such models and tools can provide insight in the root causes of performance losses.
Overall Objectives
The main goal of the Performance Plus project was to develop and demonstrate models and tools to monitor, control and test PV systems. These models and tools would then be used to optimize and enhance the performance, reliability and lifetime of commercial PV systems. Furthermore, the work focused on improved integration of PV-generated electricity into the power system through methods for short-term forecasting, testing and diagnostics, integrated energy management and storage control of thermal systems and PV system monitoring and control. The goal is to improve the competitiveness of PV on the system level.
The resulting collection of tools will be applicable to the decisive phases in the life cycle of a PV plant, namely, design, operation and maintenance. All results and models are validated with empirical data and the resulting tools are demonstrated in a relevant environment.
The project also tackled the question of energy storage. PV is increasingly becoming integrated with on-site storage and energy management systems. This is done in order to increase the sales value of the PV electricity for the owner or an aggregator acting as intermediary. Of course, it is essential for these systems to have a good control system. Suboptimal control may cause unnecessary over-sizing of storage devices or a significant drop of the overall energy conversion efficiency of the PV-energy management-storage system. Here, model-predictive control (MPC) is recommended for optimally controlling these systems.
The Performance Plus project
The Performance Plus project focused on reliable, cost-effective and highly performing photovoltaic systems. The overall objective of Performance Plus was to develop and demonstrate models and tools for monitoring, control and testing of PV systems. Means for a better integration of PV-generated electricity into the power system were provided by methods for short-term forecasting, PV system monitoring and integrated energy management and storage control.
The Performance Plus consortium looked at how to improve the PV system as a whole rather than on component level. Therefore, the PV modules and inverters were studied with focus on their operation within a system. Moreover, the Performance Plus project also studied the discrepancies between predicted and actual performance of PV systems. The consortium analyzed the way the uncertainties of all input parameters propagate throughout the PV modelling chain and affect the predicted performance. The developed models, methods and tools, aiming at reducing these uncertainties, focus on the early and reliable detection of operational performance issues by providing insight on the root causes of performance losses. This knowledge will help PV plant owners and operators to reduce the occurrence of component failures, and thus reduce their O&M costs.
The developed models, methods and tools were validated and demonstrated within the project. The Performance Plus consortium had access to a total of 25 PV systems across Europe. Where needed, the monitoring data from these plants were integrated into 3E’s monitoring and reporting platform SynaptiQ, so the developed methods and tools could be applied. Furthermore, a new type of irradiance sensor was developed within the project and prototypes were installed at five different sites across Europe for validation.
The main objective of the validation and demonstration phase was to bring the developed devices, methods and tools into practice and to ensure the quality of the results. The results of the long-term assessment of monitoring and control in view of sensing, fault detection, performance analysis, communication and effectiveness of the algorithms for energy management and storage control are presented in the public technical reports D6.2 Demonstration and validation report for sensor and monitoring system and D6.3 Demonstration and validation report for tools for field testing, both available at the project website. The results of the validation and demonstration phase serve for transforming the project results into recommendations and applicable solutions for the industry.
Available PV Plants for Validation and Demonstration
For the validation and demonstration of the models, methods and tools developed within the project, the Performance Plus consortium used monitoring data from a total of 25 PV systems across Europe. One of the installations has been in operation for more than 27 years; others have recently been installed with new PV technology. The available plants cover a wide range of installation capacity, from small residential-scale systems to large utility-scale PV plants, and are distributed over Europe as shown in Figure 1.
Figure 1 (see attached document for figures): Geographical distribution and size overview of the PV plants used for demonstration and validation activities
Project Results:
Tools and methods to help reducing the risk during design phase
The performance Plus consortium studied the effect of the uncertainties in the underlying PV modelling steps. Best practice guidelines on uncertainty in PV modelling and monitoring have been published by the project team. These guidelines will help developers and investors to evaluate more carefully the financial risks during the design phase of a PV plant. The assessment of the solar resource has been identified as the most important element in the contribution to the total uncertainty. Therefore, if appropriate, the consortium recommends combining ground measurements and satellite estimates in order to reduce the uncertainty of the solar resource assessment. The developed method called Smart Irradiation Service that combines satellite estimates and ground measured data through a kriging-of-difference methodology has been validated with data collected over 204 sites across Belgium, the Netherlands and France. The validation results show that in average, around 30% of improvement in accuracy can be achieved by this methodology (Figure 2).
Figure 2 (see attached document for figures): Normalized Root mean square error (nRMSE) of various satellite references compared with ground data from 204 weather stations in Belgium, The Netherlands and France and standard RMSE on pyranometer data
For situations where a high accuracy of the modelled output power is required, the project team has developed thermo-electric models to account for the effect of non-uniform and fast changing ambient conditions. Validation results of the developed model show that the daily estimates for average days across a period with both sunny, uniform cloudy and variable cloudy conditions, have been improved from 4% in PVSYST and 3.5% in the Sandia model to 2.5% in our new physics-based fine-grain electro-optical-thermal model. Therefore, the scientific and technical objectives for this task as described in the previous section according to the Annex I of the grant agreement where a reduction from the currently 4% to approximately 3% have been successfully reached. Moreover, results have shown that today's models fully ignore the non-uniformity across a PV module, assuming all cells to behave in an identical way. Results show that non-uniformities of up to 4-6 °C are present which means an error of 2-3% compared to the nominal average temperature values. Finally, a system scenario approach to speed-up the calculation has been introduced by the consortium. Results show that the speed up is potentially very large reducing the electrical-optical-thermal (E-O-T) model simulation from 1 day for 1 year of data down to 3 minutes (about factor 500 speedup). At the same time the daily energy yield root mean squared error (RMSE) compared to the accurate E-O-T model increases from 2.5% to 3.75% only, which is a very acceptable and interesting trade-off. When compared to PVSYST, the RMSE is even very similar (4.02 %) so no accuracy is lost then, and the speedup is still significant.
Furthermore, an innovative multi-directional irradiance sensor called ESA, that measures both the direct and indirect solar radiation was designed and built within the project. ESA’s smaller size, easier installation and operation and up to 85% of cost reduction compared with an equivalent solar irradiation station are amongst the advantages of this innovative sensor (Figure 3).
Figure 3(see attached document for figures): Comparison of the ESA sensor with its junction box (left – source: Alitec Srl) and an equivalent solar irradiance measurement station equipped with a suntracker with shaded pyranometer and pyrheliometer that measure diffuse and direct normal irradiance (right – source: University of Oldenburg)
The proper functioning of the multi-directional irradiance sensor (ESA) has been successfully demonstrated in four sites across Europe (Figure 4).
Figure 4 (see attached document for figures): Measurement setup of the ESA irradiance sensor and the reference measurement devices in (a) Lugano, (b) Oldenburg, (c) Brussels and (d) Vienna
The validation results show that the accuracy of the ESA irradiance sensor is good and comparable with second class pyranometers found in the market. Moreover, the validation results of the separate irradiation components demonstrated the ability of the ESA irradiance sensor to measure the diffuse and direct normal components of the irradiation with remarkable accuracy considering the size and price of the ESA sensor.
The ESA sensor’s innovative features result in reduced monitoring costs and maintenance efforts, thus allowing not only for small PV plants to perform solar irradiance monitoring, but also for medium to large PV plants to perform distributed rather than single-point solar irradiance monitoring. Moreover, in the field of energy management and optimal control of thermal systems, it enables the active optimization and fine-tuning of heating and/or cooling related energy use.
Moreover, the consortium developed new indoor PV module and inverter test procedures that help to characterize better the PV system components for real operating conditions and to discover quality issues before installation. In addition, thanks to the fine-grained temporal and spatial models that are able to take into account very complex environmental conditions, the overall uncertainty during design phase is reduced. Additional accelerated stress reliability tests have been developed to qualify PV inverters for real world applications under different climate conditions. Moreover, for a better understanding and estimating the long-term energy performance of PV modules, new indoor test procedures for degradation mechanisms have been introduced. The procedures will allow for determining the tendency of performance degradation mechanisms caused by potential induced degradation (PID) and dynamic load stress during the PV module’s lifetime. These new procedures continue under development within the respective working groups of the International Electrotechnical Commission (IEC).
Tools and methods to increase the performance during operation phase
The consortium has developed a methodology able to characterize the PV array through physical parameters estimated from operational data and to provide insight in the root causes of performance losses. This methodology has been called the PV Health Scan and is used to detect and diagnose faults and degradation remotely using monitoring data. The PV Health Scan methodology is illustrated in Figure 5. The method starts from closed-form relationships between regression parameters and underlying physical parameters.
Figure 5 (see attached document for figures): PV Health Scan methodology
A screenshot of some of the features of the PV Health Scan is shown in Figure 6.
Figure 6 (see attached document for figures): Screenshot of some features of the PV Health Scan
Results of the validation and demonstration phase show that the application of this method can increase the performance up to 10% thanks to the early detection of issues. Moreover, the PV Health Scan allows the systematic analysis of operational data in an efficient way, identifying how design choices and operation and maintenance (O&M) practices lead to inferior or, on the contrary, superior performance in the field. The results show how the toolbox not only detects various performance issues, but it also facilitates the root cause analysis by identifying probable root causes. Among the detected issues were string failures, inter-row shading, vegetation growth, bypass diode failures, potential induced degradation, maximum power point tracking errors, losses due to wrong inverter settings and configuration errors.
Furthermore, the project team developed a new toolbox for on-site testing of PV modules and inverters. This toolbox allows checking the components directly in the field without the need of sending them to a laboratory for testing. A screenshot of some of the features of the toolbox is shown in Figure 7.
Figure 7 (see attached document for figures): Screenshot of some features of the toolbox for on-site testing of PV modules and inverters
The toolbox for on-site testing of PV modules and inverters has been successfully demonstrated and validated in real PV plants; one of them being a very old PV plant (>30 years under operation). Results of the validation and demonstration phase show that the toolbox allows to analyse PV installations of different ages and types and to investigate the root causes behind the under-performance and energy losses.
Tools and methods to allow a better integration of PV into the power grid
The consortium developed new methods to forecast imminent irradiance changes with a sky imager that extracts the cloud distribution and their motion as well as the resulting surface shadow and solar irradiance fields. A screenshot of the short-term forecasting is shown in Figure 8.
Figure 8 (see attached document for figures): Screenshot of the short-term irradiance forecasting application to Oldenburg data using sky imaging techniques
Results of the validation show that the global horizontal irradiance (GHI) retrieved on the basis of sky images and a machine learning approach, outperforms Meteosat Second Generation (MSG) satellite derived data for the measurement site of the University of Oldenburg. RMSE values of the developed sky imager irradiance retrieval are 25% lower than for state of the art satellite models for hourly irradiance values. The developed sky imager based forecast model can provide forecasts with a temporal resolution of 1s and a spatial resolution of a few meters. Moreover, as no expensive radiometric measurement devices are required, the proposed approach reduces significantly the initial investment and operational costs.
Furthermore, in combination with the simulation models developed within the project, the accuracy of PV energy yield nowcasting methods has been improved. The validation results show that for a horizon of five minutes, the developed PV energy yield nowcasting method combining sky imaging with simulation models outperforms the persistence model. Moreover, the accuracy of the proposed model is in line with state-of-the-art energy yield forecasts, but with the added value that the irradiation-to-energy error is already included. Additionally, promising results have been obtained from the validation of the modelling of coupled cell-inverter thermal behaviour.
Furthermore, the project team focussed on minimizing the operational costs of installations by means of optimal controllers. The consortium has developed a Model Predictive Control (MPC) framework that allows harnessing flexibility from any controllable load and enables a better integration of PV plants into the power grid. In the MPC framework, the system is controlled using control inputs that are found by optimizing the model-predicted future behaviour of the system as shown in Figure 9.
Figure 9 (see attached document for figures): Overview of the Model Predictive Control (MPC) framework
The developed method for energy management and storage control has been demonstrated through three use cases comprising PV, controllable thermal energy resources and energy storage. For the heating case, results show that the self-consumption rate can be increased from 54% to 74% by minimizing the system’s operational cost for building energy use and PV production compared to the building energy use only. This is a 36% increase in self-consumption. At the same time, the self-production is increased with 29%. These results from the implementation of the MCP in an office building in Brussels show that the use of model predictive controllers (MPC) during the winter in Belgium allows saving up to 30% of the heating energy costs compared to the initially used rule based controller. For the cooling case, the simulation results show that self-consumption rate can be increased from 77% (set point tracking reference) or 54% (optimal energy use controller) to 100%. This is a 29% or 85% increase respectively. Moreover, while improving the self-consumption, also the self-production was increased and also the total operational cost was reduced with 30%.
Potential Impact:
The Performance Plus project has developed models, methods and tools to optimize and enhance the performance, reliability and lifetime of commercial PV systems beyond the state-of-the-art. The results were validated and demonstrated with more than 50 MW of empirical data from a total of 25 PV systems across Europe. The results of the validation and demonstration phase are publicly available on the project website and have been presented in both scientific conferences and in public workshops organized by the consortium. Novel methods have been introduced and validated for advanced PV module modelling, short-term forecasting, testing and diagnostics, advanced PV system monitoring techniques, and integrated energy management and storage control. Some of these project results are already being transformed into practically applicable solutions and proven marketable products.
The following models, methods and tools to assess and optimize the performance of PV systems have been developed by the Performance Plus consortium:
Tools and methods to reduce the risk during design phase
• Smart Irradiation Service (SIS)
• Multi-directional irradiance sensor (ESA sensor)
• Fine-grained models
• Indoor PV module and inverter test procedures
Tools and methods to increase performance during operation phase by detecting and diagnosing failures
• PV Health Scan and Solar Sensor Check
• Toolbox for on-site testing of PV modules and inverters
Tools and methods to integrate PV into power system by optimally managing and controlling the energy
• Sky imager
• PV energy yield nowcasting combining sky imaging with simulation models
• Model Predictive Control (MPC) for optimal energy management and storage control
The pursued impacts grouped by target group are listed below.
Policy makers
• Cost Reduction: Investment costs, inverter lifetime, module power over time, system performance ratio
• System Integration: Private self-consumption, energy management/storage control & communication for grid support, short-term production forecast.
Utility sector (Grid operators, energy suppliers, energy service companies)
Improved economics of integrated PV with storage and energy management systems. Smart grid and power system integration.
PV system integrators, plant operators, energy aggregators, financing institutions
Reduced investment risk, PV forecast quality, gains from energy management & storage control, improved PV system performance and reliability, reduced occurrence of false alarms & reliable detection of real faults.
PV module, inverter and other component manufacturers
Improved PV system & component performance and reliability, compatible communication interfaces.
Engineering, design, monitoring and forecasting service providers
Models improving existing design & monitoring tools, higher confidence in energy yield predictions,
PV forecast quality, reduced occurrence of false alarms & reliable detection of real faults.
Research & higher education
Progress beyond the state of the art in the core technologies: PV system & component modelling, forecasting, energy modelling & optimisation, MPC sensing, pattern recognition, data mining, inverter control and diagnostic testing.
Media, civil society, interested public
Foster dialogue and debate on strategic impacts: LCOE reduction and system integration/ grids, progress towards KPIs of the SEII.
Dissemination Activities
Over the project duration, the consortium partners wrote 30 scientific papers of which 7 were published in peer-reviewed journals. All the papers were presented at various scientific conferences through Europe. Besides the scientific publications, the consortium members were invited to present the project results in different occasions. In total, XX invited presentation were given by the project members.
Furthermore, two project workshops were organized by the consortium to disseminate the project findings, lessons learned and recommendations.
The first project workshop took place on Wednesday 19th of November 2014 in Brussels, Belgium (Figure 10). This workshop counted with the participation of the UG members and was focused on exploring how the new models and tools that were being developed in the project can improve the performance of PV systems.
Figure 10 (see attached document for figures): First Performance Plus workshop
The second project workshop was held together with another FP7 project (PVCROPS) and took place on Tuesday 6th of October 2015 in Brussels, Belgium (Figure 11). This closing event was focused on the dissemination of the final results of the demonstration and validation phase of the developed devices, methods and tools.
Figure 11 (see attached document for figures): Performance Plus closing event
All presentations of both events are publically available at the project website under http://perfplus.eu/news. Furthermore, the consortium prepared an infographic (Figure 12) for the closing event on October 2015. The infographic summarizes the project outcomes and presents quantitative results.
Figure 12 (see attached document for figures): Performance Plus project infographic
Furthermore, three press releases targeted to the civil society were launched during the duration of the project. All the press releases were disseminated through different communication channels e.g. Figure 13 shows one of the press releases, leading to an article in the Sun & Wind Energy magazine for renewable energies in March 2014. Moreover, all press releases were also disseminated through the project website (http://perfplus.eu/news).
Figure 13 (see attached document for figures): Performance Plus press release published in Sun & Wind Energy magazine
Exploitation of Results
The project partners are able to exploit the project results in the following ways:
3E
• PV Health Scan: 3E is planning to exploit the PV health scan method in a first stage as a service linked to 3E’s monitoring portal SynaptiQ. In the longer term, this can also be offered as a consultancy service.
• Smart Irradiation Service (SIS): 3E is already applying this method for consultancy services as a reference irradiation for projects with no existing or low quality irradiation measurements.
• Model Predictive Control (MPC) for optimal energy management and control: 3E plans further collaboration with KU Leuven towards offering commercial MPC services.
Alitec
• Multi-directional irradiance sensor (ESA sensor): Alitec will offer the ESA sensor and provide technical assistance to EPC and O&M contractors.
IMEC
• Tools for detailed thermal optical electrical energy yield modelling of PV modules: IMEC will continue with the improvement of these models. Moreover, the following customer segments are targeted for their further application:
- PV rooftop or plant installers/aggregators and solar car or transport providers: training and transfer of methodologies and tools in prototype form;
- PV energy yield model providers: in depth transfer of methodologies and tools in “white-box” mode. They will provide similar services as IMEC to customer segments described above, but in a fully commercial model. 3E is considered as a prime partner to start with.
• PV energy yield nowcasting combining sky imaging with simulation models: this is a joint result shared with the University of Oldenburg. It has been agreed by both partners that this results can be used in the future by both sides without accounting so they provide a cross-license towards each other for this information and the usage in demonstrations. Both partners plan to extend this cooperation beyond the project to further build on the obtained results.
• Thermal model for mutual impact of the local DC-AC convertor and PV module in a PV string setup: this is a joint result shared with AIT. It has been agreed by both partners that both the instantiated model and the experimental results obtained are considered as public knowledge (a joint journal paper is foreseen). This knowledge can be used in the future by both sides without accounting. However, both parties will only use this while always mutually citing each other’s contribution.
KU Leuven
• Model Predictive Control (MPC) for optimal energy management and control: KU Leuven will continue the development of MPC. Moreover, negotiations for future cooperation have started between KU Leuven and 3E. Both partners plan to extend their cooperation beyond the project to further build on the obtained results.
SUPSI
• Indoor PV module test procedures:
• Toolbox for on-site testing of PV modules and inverters: SUPSI and AIT have collaborated into the development of new procedures and tools for field testing on PV plants. Both the instantiated model and the experimental results will be used in the future by both sides without accounting. However, both parties will only use this while always mutually citing each other’s contribution.
UOL
• Sky imager: The University of Oldenburg will continue with the improvement of the solar irradiation forecasting based on Sky imaging.
• PV energy yield nowcasting combining sky imaging with simulation models: This is a joint result shared with IMEC. It has been agreed by both partners that this results can be used in the future by both sides without accounting so they provide a cross-license towards each other for this information and the usage in demonstrations. Both partners plan to extend this cooperation beyond the project to further build on the obtained results.
AIT
• Switch based PV inverter models: AIT is planning to offer this as a consultancy service for inverter manufacturers.
• PV inverter average models: AIT will use these accurate PV inverter average models to predict possible revenue of PV plants in a design stage based on site related mission profiles. Furthermore, the models can also be used to calculate the possible revenue from running PV plants and help to identify and analyse deviations from this maximum value. AIT will target PV power plant owners as well as consultants to analyse PV plants using this model. Furthermore, due to the low computing resources needed, private users would also be able to analyse their expected revenue based on a best-case calculation.
• PV inverter reliability and robustness tests: AIT will offer these tests as a consultancy service for inverter manufacturers.
• PV inverter lifetime estimation: AIT will target with this service mainly PV inverter manufacturers. Furthermore, PV site planners could also use this service for non-standard fields of application to evaluate the risk of failure during the time of use.
• PV “Golden Inverter” tool: AIT will target PV site owners/consultants/service providers to help to monitor PV inverter performance using either internal inverter readings or measurement devices.
• Toolbox for on-site testing of PV modules and inverters: AIT and SUPSI have collaborated into the development of new procedures and tools for field testing on PV plants. Both the instantiated model and the experimental results will be used in the future by both sides without accounting. However, both parties will only use this while always mutually citing each other’s contribution.
• Thermal model for mutual impact of the local DC-AC convertor and PV module in a PV string setup: this is a joint result shared with IMEC. It has been agreed by both partners that both the instantiated model and the experimental results obtained are considered as public knowledge (a joint journal paper is foreseen). This knowledge can be used in the future by both sides without accounting. However, both parties will only use this while always mutually citing each other’s contribution.
List of Websites:
www.perfplus.eu
The Performance Plus project was structured in two main phases. In the first phase, a series of devices, models, methods and tools to assess and optimize the performance and lifetime of photovoltaic (PV) installations have been developed. In the second phase, the results have been validated and demonstrated thanks to a User Group composed of large portfolio owners and strong EPC and O&M contractors that gave access to 25 PV systems across Europe making a total of more than 50 MW of installed capacity. In a nutshell, the final results of the project are:
Tools and methods to help reducing the risk during design phase
A method to combine satellite derived irradiation and ground measured data called Smart Irradiation Service has been developed. Up to 30% improvement in accuracy has been achieved by applying this method. Moreover, an innovative multi-directional irradiance sensor called ESA, that measures both the direct and indirect solar radiation was designed and built within the project. ESA’s smaller size, easier installation and operation and up to 85% of cost reduction compared with an equivalent solar irradiation station are amongst the advantages of this innovative sensor.
The new indoor PV module and inverter test procedures developed by the consortium help to characterize better the PV system components for real operating conditions and to discover quality issues before installation. In addition, thanks to the fine-grained temporal and spatial models that are able to take into account very complex environmental conditions, the overall uncertainty during design phase is reduced.
Tools and methods to increase the performance during operation phase
The consortium has developed a method to detect and diagnose faults and degradation remotely using monitoring data. The application of this method called PV Health Scan can increase the performance up to 10% thanks to the early detection of issues. Furthermore, the new toolbox for on-site testing of PV modules and inverters developed by the consortium allows to check the components directly on-site without the need of sending them to a laboratory for testing.
Tools and methods to allow a better integration of PV into the power grid
The consortium developed new methods to forecast imminent irradiance changes thanks to a sky imager that extracts the cloud distribution and their motion as well as the resulting surface shadow and solar irradiance fields. In combination with the fine-grain simulation models developed within the project, the accuracy of PV energy yield nowcasting methods is significantly improved. Moreover, the application of a Model Predictive Control (MPC) framework that allows harnessing flexibility from any controllable load enables a better integration of PV plants into the power grid. As demonstrated by the consortium, the application of this method can save up to 30% on heating energy costs during winter in Belgium.
The Performance Plus consortium presented the final results of this three years project to an audience of industry and researchers during a closing event, held in Brussels on the 6th of October 2015. The presentations given during this closing event and all public results are available on the project website: www.perfplus.eu
Project Context and Objectives:
Beyond Improving Components: System Level Optimization
The PV industry is today at an interesting point in its history. To ensure a continuous decrease of the costs linked to electricity from PV, the prices of modules, inverters and balance of system (BOS) components have to be further decreased. This has to be accomplished, while performance, functionality, reliability and lifetime on the component and system level need to be increased. Needless to say, the industry has made large technological progress with PV cells, modules and inverters in terms of costs and reliability. However, from an integrated perspective, PV system performance emerges from, but is not limited to, the performance of the components. Therefore, in order to improve the performance of PV systems, we need to look at how we can improve the system as a whole.
Limitations during Design Phase: Predicted vs Real Performance
There are several PV performance modelling software packages available on the market, specially developed to predict the amount of energy that a PV system can produce. The predicted performance from these models can differ significantly between each other and from actual performance from real systems. One of the main goals of this work was to better understand these discrepancies between predicted and actual performance of PV systems.
The large amount of input parameters like irradiance, temperature, array orientation, module and inverter performance, user-defined values for additional losses such as soiling, mismatching, cabling, etc. have inherent uncertainties. These uncertainties have to be properly taken into account, as their correct quantification is essential for evaluating the financial risk of PV investments.
Improving Operation & Maintenance: Early and Reliable Detection of Operational Performance Issues
For PV professionals, the early detection and diagnosis of faults is of the utmost importance to obtain and maintain high energy yield of PV systems. Moreover, the timely remediation of faults not only restores the production promptly but also avoids the occurrence of additional component failures and leads to reduction of costs for operation and maintenance (O&M). However, this early and reliable detection of issues requires accurate models of the expected behaviour of the well-performing PV plant. In practice, tools are required to help PV plant owners and operators reducing the occurrence of component failures and thus reducing O&M costs. Ideally, such models and tools can provide insight in the root causes of performance losses.
Overall Objectives
The main goal of the Performance Plus project was to develop and demonstrate models and tools to monitor, control and test PV systems. These models and tools would then be used to optimize and enhance the performance, reliability and lifetime of commercial PV systems. Furthermore, the work focused on improved integration of PV-generated electricity into the power system through methods for short-term forecasting, testing and diagnostics, integrated energy management and storage control of thermal systems and PV system monitoring and control. The goal is to improve the competitiveness of PV on the system level.
The resulting collection of tools will be applicable to the decisive phases in the life cycle of a PV plant, namely, design, operation and maintenance. All results and models are validated with empirical data and the resulting tools are demonstrated in a relevant environment.
The project also tackled the question of energy storage. PV is increasingly becoming integrated with on-site storage and energy management systems. This is done in order to increase the sales value of the PV electricity for the owner or an aggregator acting as intermediary. Of course, it is essential for these systems to have a good control system. Suboptimal control may cause unnecessary over-sizing of storage devices or a significant drop of the overall energy conversion efficiency of the PV-energy management-storage system. Here, model-predictive control (MPC) is recommended for optimally controlling these systems.
The Performance Plus project
The Performance Plus project focused on reliable, cost-effective and highly performing photovoltaic systems. The overall objective of Performance Plus was to develop and demonstrate models and tools for monitoring, control and testing of PV systems. Means for a better integration of PV-generated electricity into the power system were provided by methods for short-term forecasting, PV system monitoring and integrated energy management and storage control.
The Performance Plus consortium looked at how to improve the PV system as a whole rather than on component level. Therefore, the PV modules and inverters were studied with focus on their operation within a system. Moreover, the Performance Plus project also studied the discrepancies between predicted and actual performance of PV systems. The consortium analyzed the way the uncertainties of all input parameters propagate throughout the PV modelling chain and affect the predicted performance. The developed models, methods and tools, aiming at reducing these uncertainties, focus on the early and reliable detection of operational performance issues by providing insight on the root causes of performance losses. This knowledge will help PV plant owners and operators to reduce the occurrence of component failures, and thus reduce their O&M costs.
The developed models, methods and tools were validated and demonstrated within the project. The Performance Plus consortium had access to a total of 25 PV systems across Europe. Where needed, the monitoring data from these plants were integrated into 3E’s monitoring and reporting platform SynaptiQ, so the developed methods and tools could be applied. Furthermore, a new type of irradiance sensor was developed within the project and prototypes were installed at five different sites across Europe for validation.
The main objective of the validation and demonstration phase was to bring the developed devices, methods and tools into practice and to ensure the quality of the results. The results of the long-term assessment of monitoring and control in view of sensing, fault detection, performance analysis, communication and effectiveness of the algorithms for energy management and storage control are presented in the public technical reports D6.2 Demonstration and validation report for sensor and monitoring system and D6.3 Demonstration and validation report for tools for field testing, both available at the project website. The results of the validation and demonstration phase serve for transforming the project results into recommendations and applicable solutions for the industry.
Available PV Plants for Validation and Demonstration
For the validation and demonstration of the models, methods and tools developed within the project, the Performance Plus consortium used monitoring data from a total of 25 PV systems across Europe. One of the installations has been in operation for more than 27 years; others have recently been installed with new PV technology. The available plants cover a wide range of installation capacity, from small residential-scale systems to large utility-scale PV plants, and are distributed over Europe as shown in Figure 1.
Figure 1 (see attached document for figures): Geographical distribution and size overview of the PV plants used for demonstration and validation activities
Project Results:
Tools and methods to help reducing the risk during design phase
The performance Plus consortium studied the effect of the uncertainties in the underlying PV modelling steps. Best practice guidelines on uncertainty in PV modelling and monitoring have been published by the project team. These guidelines will help developers and investors to evaluate more carefully the financial risks during the design phase of a PV plant. The assessment of the solar resource has been identified as the most important element in the contribution to the total uncertainty. Therefore, if appropriate, the consortium recommends combining ground measurements and satellite estimates in order to reduce the uncertainty of the solar resource assessment. The developed method called Smart Irradiation Service that combines satellite estimates and ground measured data through a kriging-of-difference methodology has been validated with data collected over 204 sites across Belgium, the Netherlands and France. The validation results show that in average, around 30% of improvement in accuracy can be achieved by this methodology (Figure 2).
Figure 2 (see attached document for figures): Normalized Root mean square error (nRMSE) of various satellite references compared with ground data from 204 weather stations in Belgium, The Netherlands and France and standard RMSE on pyranometer data
For situations where a high accuracy of the modelled output power is required, the project team has developed thermo-electric models to account for the effect of non-uniform and fast changing ambient conditions. Validation results of the developed model show that the daily estimates for average days across a period with both sunny, uniform cloudy and variable cloudy conditions, have been improved from 4% in PVSYST and 3.5% in the Sandia model to 2.5% in our new physics-based fine-grain electro-optical-thermal model. Therefore, the scientific and technical objectives for this task as described in the previous section according to the Annex I of the grant agreement where a reduction from the currently 4% to approximately 3% have been successfully reached. Moreover, results have shown that today's models fully ignore the non-uniformity across a PV module, assuming all cells to behave in an identical way. Results show that non-uniformities of up to 4-6 °C are present which means an error of 2-3% compared to the nominal average temperature values. Finally, a system scenario approach to speed-up the calculation has been introduced by the consortium. Results show that the speed up is potentially very large reducing the electrical-optical-thermal (E-O-T) model simulation from 1 day for 1 year of data down to 3 minutes (about factor 500 speedup). At the same time the daily energy yield root mean squared error (RMSE) compared to the accurate E-O-T model increases from 2.5% to 3.75% only, which is a very acceptable and interesting trade-off. When compared to PVSYST, the RMSE is even very similar (4.02 %) so no accuracy is lost then, and the speedup is still significant.
Furthermore, an innovative multi-directional irradiance sensor called ESA, that measures both the direct and indirect solar radiation was designed and built within the project. ESA’s smaller size, easier installation and operation and up to 85% of cost reduction compared with an equivalent solar irradiation station are amongst the advantages of this innovative sensor (Figure 3).
Figure 3(see attached document for figures): Comparison of the ESA sensor with its junction box (left – source: Alitec Srl) and an equivalent solar irradiance measurement station equipped with a suntracker with shaded pyranometer and pyrheliometer that measure diffuse and direct normal irradiance (right – source: University of Oldenburg)
The proper functioning of the multi-directional irradiance sensor (ESA) has been successfully demonstrated in four sites across Europe (Figure 4).
Figure 4 (see attached document for figures): Measurement setup of the ESA irradiance sensor and the reference measurement devices in (a) Lugano, (b) Oldenburg, (c) Brussels and (d) Vienna
The validation results show that the accuracy of the ESA irradiance sensor is good and comparable with second class pyranometers found in the market. Moreover, the validation results of the separate irradiation components demonstrated the ability of the ESA irradiance sensor to measure the diffuse and direct normal components of the irradiation with remarkable accuracy considering the size and price of the ESA sensor.
The ESA sensor’s innovative features result in reduced monitoring costs and maintenance efforts, thus allowing not only for small PV plants to perform solar irradiance monitoring, but also for medium to large PV plants to perform distributed rather than single-point solar irradiance monitoring. Moreover, in the field of energy management and optimal control of thermal systems, it enables the active optimization and fine-tuning of heating and/or cooling related energy use.
Moreover, the consortium developed new indoor PV module and inverter test procedures that help to characterize better the PV system components for real operating conditions and to discover quality issues before installation. In addition, thanks to the fine-grained temporal and spatial models that are able to take into account very complex environmental conditions, the overall uncertainty during design phase is reduced. Additional accelerated stress reliability tests have been developed to qualify PV inverters for real world applications under different climate conditions. Moreover, for a better understanding and estimating the long-term energy performance of PV modules, new indoor test procedures for degradation mechanisms have been introduced. The procedures will allow for determining the tendency of performance degradation mechanisms caused by potential induced degradation (PID) and dynamic load stress during the PV module’s lifetime. These new procedures continue under development within the respective working groups of the International Electrotechnical Commission (IEC).
Tools and methods to increase the performance during operation phase
The consortium has developed a methodology able to characterize the PV array through physical parameters estimated from operational data and to provide insight in the root causes of performance losses. This methodology has been called the PV Health Scan and is used to detect and diagnose faults and degradation remotely using monitoring data. The PV Health Scan methodology is illustrated in Figure 5. The method starts from closed-form relationships between regression parameters and underlying physical parameters.
Figure 5 (see attached document for figures): PV Health Scan methodology
A screenshot of some of the features of the PV Health Scan is shown in Figure 6.
Figure 6 (see attached document for figures): Screenshot of some features of the PV Health Scan
Results of the validation and demonstration phase show that the application of this method can increase the performance up to 10% thanks to the early detection of issues. Moreover, the PV Health Scan allows the systematic analysis of operational data in an efficient way, identifying how design choices and operation and maintenance (O&M) practices lead to inferior or, on the contrary, superior performance in the field. The results show how the toolbox not only detects various performance issues, but it also facilitates the root cause analysis by identifying probable root causes. Among the detected issues were string failures, inter-row shading, vegetation growth, bypass diode failures, potential induced degradation, maximum power point tracking errors, losses due to wrong inverter settings and configuration errors.
Furthermore, the project team developed a new toolbox for on-site testing of PV modules and inverters. This toolbox allows checking the components directly in the field without the need of sending them to a laboratory for testing. A screenshot of some of the features of the toolbox is shown in Figure 7.
Figure 7 (see attached document for figures): Screenshot of some features of the toolbox for on-site testing of PV modules and inverters
The toolbox for on-site testing of PV modules and inverters has been successfully demonstrated and validated in real PV plants; one of them being a very old PV plant (>30 years under operation). Results of the validation and demonstration phase show that the toolbox allows to analyse PV installations of different ages and types and to investigate the root causes behind the under-performance and energy losses.
Tools and methods to allow a better integration of PV into the power grid
The consortium developed new methods to forecast imminent irradiance changes with a sky imager that extracts the cloud distribution and their motion as well as the resulting surface shadow and solar irradiance fields. A screenshot of the short-term forecasting is shown in Figure 8.
Figure 8 (see attached document for figures): Screenshot of the short-term irradiance forecasting application to Oldenburg data using sky imaging techniques
Results of the validation show that the global horizontal irradiance (GHI) retrieved on the basis of sky images and a machine learning approach, outperforms Meteosat Second Generation (MSG) satellite derived data for the measurement site of the University of Oldenburg. RMSE values of the developed sky imager irradiance retrieval are 25% lower than for state of the art satellite models for hourly irradiance values. The developed sky imager based forecast model can provide forecasts with a temporal resolution of 1s and a spatial resolution of a few meters. Moreover, as no expensive radiometric measurement devices are required, the proposed approach reduces significantly the initial investment and operational costs.
Furthermore, in combination with the simulation models developed within the project, the accuracy of PV energy yield nowcasting methods has been improved. The validation results show that for a horizon of five minutes, the developed PV energy yield nowcasting method combining sky imaging with simulation models outperforms the persistence model. Moreover, the accuracy of the proposed model is in line with state-of-the-art energy yield forecasts, but with the added value that the irradiation-to-energy error is already included. Additionally, promising results have been obtained from the validation of the modelling of coupled cell-inverter thermal behaviour.
Furthermore, the project team focussed on minimizing the operational costs of installations by means of optimal controllers. The consortium has developed a Model Predictive Control (MPC) framework that allows harnessing flexibility from any controllable load and enables a better integration of PV plants into the power grid. In the MPC framework, the system is controlled using control inputs that are found by optimizing the model-predicted future behaviour of the system as shown in Figure 9.
Figure 9 (see attached document for figures): Overview of the Model Predictive Control (MPC) framework
The developed method for energy management and storage control has been demonstrated through three use cases comprising PV, controllable thermal energy resources and energy storage. For the heating case, results show that the self-consumption rate can be increased from 54% to 74% by minimizing the system’s operational cost for building energy use and PV production compared to the building energy use only. This is a 36% increase in self-consumption. At the same time, the self-production is increased with 29%. These results from the implementation of the MCP in an office building in Brussels show that the use of model predictive controllers (MPC) during the winter in Belgium allows saving up to 30% of the heating energy costs compared to the initially used rule based controller. For the cooling case, the simulation results show that self-consumption rate can be increased from 77% (set point tracking reference) or 54% (optimal energy use controller) to 100%. This is a 29% or 85% increase respectively. Moreover, while improving the self-consumption, also the self-production was increased and also the total operational cost was reduced with 30%.
Potential Impact:
The Performance Plus project has developed models, methods and tools to optimize and enhance the performance, reliability and lifetime of commercial PV systems beyond the state-of-the-art. The results were validated and demonstrated with more than 50 MW of empirical data from a total of 25 PV systems across Europe. The results of the validation and demonstration phase are publicly available on the project website and have been presented in both scientific conferences and in public workshops organized by the consortium. Novel methods have been introduced and validated for advanced PV module modelling, short-term forecasting, testing and diagnostics, advanced PV system monitoring techniques, and integrated energy management and storage control. Some of these project results are already being transformed into practically applicable solutions and proven marketable products.
The following models, methods and tools to assess and optimize the performance of PV systems have been developed by the Performance Plus consortium:
Tools and methods to reduce the risk during design phase
• Smart Irradiation Service (SIS)
• Multi-directional irradiance sensor (ESA sensor)
• Fine-grained models
• Indoor PV module and inverter test procedures
Tools and methods to increase performance during operation phase by detecting and diagnosing failures
• PV Health Scan and Solar Sensor Check
• Toolbox for on-site testing of PV modules and inverters
Tools and methods to integrate PV into power system by optimally managing and controlling the energy
• Sky imager
• PV energy yield nowcasting combining sky imaging with simulation models
• Model Predictive Control (MPC) for optimal energy management and storage control
The pursued impacts grouped by target group are listed below.
Policy makers
• Cost Reduction: Investment costs, inverter lifetime, module power over time, system performance ratio
• System Integration: Private self-consumption, energy management/storage control & communication for grid support, short-term production forecast.
Utility sector (Grid operators, energy suppliers, energy service companies)
Improved economics of integrated PV with storage and energy management systems. Smart grid and power system integration.
PV system integrators, plant operators, energy aggregators, financing institutions
Reduced investment risk, PV forecast quality, gains from energy management & storage control, improved PV system performance and reliability, reduced occurrence of false alarms & reliable detection of real faults.
PV module, inverter and other component manufacturers
Improved PV system & component performance and reliability, compatible communication interfaces.
Engineering, design, monitoring and forecasting service providers
Models improving existing design & monitoring tools, higher confidence in energy yield predictions,
PV forecast quality, reduced occurrence of false alarms & reliable detection of real faults.
Research & higher education
Progress beyond the state of the art in the core technologies: PV system & component modelling, forecasting, energy modelling & optimisation, MPC sensing, pattern recognition, data mining, inverter control and diagnostic testing.
Media, civil society, interested public
Foster dialogue and debate on strategic impacts: LCOE reduction and system integration/ grids, progress towards KPIs of the SEII.
Dissemination Activities
Over the project duration, the consortium partners wrote 30 scientific papers of which 7 were published in peer-reviewed journals. All the papers were presented at various scientific conferences through Europe. Besides the scientific publications, the consortium members were invited to present the project results in different occasions. In total, XX invited presentation were given by the project members.
Furthermore, two project workshops were organized by the consortium to disseminate the project findings, lessons learned and recommendations.
The first project workshop took place on Wednesday 19th of November 2014 in Brussels, Belgium (Figure 10). This workshop counted with the participation of the UG members and was focused on exploring how the new models and tools that were being developed in the project can improve the performance of PV systems.
Figure 10 (see attached document for figures): First Performance Plus workshop
The second project workshop was held together with another FP7 project (PVCROPS) and took place on Tuesday 6th of October 2015 in Brussels, Belgium (Figure 11). This closing event was focused on the dissemination of the final results of the demonstration and validation phase of the developed devices, methods and tools.
Figure 11 (see attached document for figures): Performance Plus closing event
All presentations of both events are publically available at the project website under http://perfplus.eu/news. Furthermore, the consortium prepared an infographic (Figure 12) for the closing event on October 2015. The infographic summarizes the project outcomes and presents quantitative results.
Figure 12 (see attached document for figures): Performance Plus project infographic
Furthermore, three press releases targeted to the civil society were launched during the duration of the project. All the press releases were disseminated through different communication channels e.g. Figure 13 shows one of the press releases, leading to an article in the Sun & Wind Energy magazine for renewable energies in March 2014. Moreover, all press releases were also disseminated through the project website (http://perfplus.eu/news).
Figure 13 (see attached document for figures): Performance Plus press release published in Sun & Wind Energy magazine
Exploitation of Results
The project partners are able to exploit the project results in the following ways:
3E
• PV Health Scan: 3E is planning to exploit the PV health scan method in a first stage as a service linked to 3E’s monitoring portal SynaptiQ. In the longer term, this can also be offered as a consultancy service.
• Smart Irradiation Service (SIS): 3E is already applying this method for consultancy services as a reference irradiation for projects with no existing or low quality irradiation measurements.
• Model Predictive Control (MPC) for optimal energy management and control: 3E plans further collaboration with KU Leuven towards offering commercial MPC services.
Alitec
• Multi-directional irradiance sensor (ESA sensor): Alitec will offer the ESA sensor and provide technical assistance to EPC and O&M contractors.
IMEC
• Tools for detailed thermal optical electrical energy yield modelling of PV modules: IMEC will continue with the improvement of these models. Moreover, the following customer segments are targeted for their further application:
- PV rooftop or plant installers/aggregators and solar car or transport providers: training and transfer of methodologies and tools in prototype form;
- PV energy yield model providers: in depth transfer of methodologies and tools in “white-box” mode. They will provide similar services as IMEC to customer segments described above, but in a fully commercial model. 3E is considered as a prime partner to start with.
• PV energy yield nowcasting combining sky imaging with simulation models: this is a joint result shared with the University of Oldenburg. It has been agreed by both partners that this results can be used in the future by both sides without accounting so they provide a cross-license towards each other for this information and the usage in demonstrations. Both partners plan to extend this cooperation beyond the project to further build on the obtained results.
• Thermal model for mutual impact of the local DC-AC convertor and PV module in a PV string setup: this is a joint result shared with AIT. It has been agreed by both partners that both the instantiated model and the experimental results obtained are considered as public knowledge (a joint journal paper is foreseen). This knowledge can be used in the future by both sides without accounting. However, both parties will only use this while always mutually citing each other’s contribution.
KU Leuven
• Model Predictive Control (MPC) for optimal energy management and control: KU Leuven will continue the development of MPC. Moreover, negotiations for future cooperation have started between KU Leuven and 3E. Both partners plan to extend their cooperation beyond the project to further build on the obtained results.
SUPSI
• Indoor PV module test procedures:
• Toolbox for on-site testing of PV modules and inverters: SUPSI and AIT have collaborated into the development of new procedures and tools for field testing on PV plants. Both the instantiated model and the experimental results will be used in the future by both sides without accounting. However, both parties will only use this while always mutually citing each other’s contribution.
UOL
• Sky imager: The University of Oldenburg will continue with the improvement of the solar irradiation forecasting based on Sky imaging.
• PV energy yield nowcasting combining sky imaging with simulation models: This is a joint result shared with IMEC. It has been agreed by both partners that this results can be used in the future by both sides without accounting so they provide a cross-license towards each other for this information and the usage in demonstrations. Both partners plan to extend this cooperation beyond the project to further build on the obtained results.
AIT
• Switch based PV inverter models: AIT is planning to offer this as a consultancy service for inverter manufacturers.
• PV inverter average models: AIT will use these accurate PV inverter average models to predict possible revenue of PV plants in a design stage based on site related mission profiles. Furthermore, the models can also be used to calculate the possible revenue from running PV plants and help to identify and analyse deviations from this maximum value. AIT will target PV power plant owners as well as consultants to analyse PV plants using this model. Furthermore, due to the low computing resources needed, private users would also be able to analyse their expected revenue based on a best-case calculation.
• PV inverter reliability and robustness tests: AIT will offer these tests as a consultancy service for inverter manufacturers.
• PV inverter lifetime estimation: AIT will target with this service mainly PV inverter manufacturers. Furthermore, PV site planners could also use this service for non-standard fields of application to evaluate the risk of failure during the time of use.
• PV “Golden Inverter” tool: AIT will target PV site owners/consultants/service providers to help to monitor PV inverter performance using either internal inverter readings or measurement devices.
• Toolbox for on-site testing of PV modules and inverters: AIT and SUPSI have collaborated into the development of new procedures and tools for field testing on PV plants. Both the instantiated model and the experimental results will be used in the future by both sides without accounting. However, both parties will only use this while always mutually citing each other’s contribution.
• Thermal model for mutual impact of the local DC-AC convertor and PV module in a PV string setup: this is a joint result shared with IMEC. It has been agreed by both partners that both the instantiated model and the experimental results obtained are considered as public knowledge (a joint journal paper is foreseen). This knowledge can be used in the future by both sides without accounting. However, both parties will only use this while always mutually citing each other’s contribution.
List of Websites:
www.perfplus.eu
