Final Report Summary - E-HUB (Energy-Hub for residential and commercial districts and transport)
Executive Summary:
Due to finite stocks of fossil fuels and the effect of greenhouse gas emissions, the amount of renewable energy from wind, biomass and solar energy etc. must strongly increase over present levels. However, due to the fluctuating nature of renewable energy supply, application of short term and long term energy storage and intelligent energy management systems are essential to match demand and supply of energy.
Thermal storage is sometimes called ‘the holy grail’ of energy neutral buildings. The consortium made good progress in improving several types of thermal storage, in particular on Thermo-Active foundations, Thermo-Chemical storage and distributed thermal storage.
The Multi Commodity Matcher (MCM) control algorithm developed in the E-hub project is able to match the supply and demand of electricity and heat simultaneously. The MCM control algorithm was used in three types of applications:
1. A simulation tool was developed in the project to carry out simulations of a virtual application of an advanced Energy Management System in five case studies: the districts of Amsterdam (NL), Freiburg (D), Bergamo (It), Leuven (B) and Dalian (China).
The main conclusion is that the MCM was able to accommodate the introduction of technologies based on Renewable Energy Sources (RES) and/or Recovered Energy (REC) within the district (GREEN/Low Carbon scenario). It resulted in an important decrease in primary energy use and CO2 emissions and leads to more beneficial cash-flows for the studied cases. Moreover, by introducing smart capabilities (in the “SMART” scenario) extra savings in costs can be realized. In the SMART scenario the environmental impact may decrease or increase depending on the business case selected.
2. The MCM was used to control the operation of several cogeneration units in a real lab environment, under conflicting demand profiles of heat and electricity.
The main conclusion is that both in tests with and without thermal storage, the MCM shows robustness in control, coming close to the best thermo-economic solution. The latter was found in a simulation with the “ECoMP” optimisation software.
3. An ‘electricity only’ version of the MCM was applied in a full scale demonstration in the district of Tweewaters in Leuven, Belgium.
The main conclusion is that the consortium set a best practice with the full scale demonstration of a high quality building equipped with a smart energy management system.
In the frame of a ‘Joint Exploitation Agreement, the simulation tool can be used by consortium members to continue to provide consulting services e.g. to municipalities in configuring energy neutral/energy efficient districts.
An important element is the acceptance of such an advanced energy system by stakeholders such as DSOs (Distribution System Operators), BRPs (Balancing Responsible Parties) and end users. Therefore, a number of new business models and service concepts, most of them based on the concept of ‘flexibility of demand’ were developed and applied in the case study simulations.
Project Context and Objectives:
INTRODUCTION
To achieve low energy or even energy neutral districts, the share of renewable energy must increase drastically over present levels. However, accommodating a large supplier of renewable energy, such as a wind farm, in the existing energy infrastructure is complicated by the fluctuating character of the energy supply with the result that renewable energy supply may be either too large or too small to cover the instantaneous demand. Both smart energy management systems and energy storage are essential to meet this challenge.
The objective of the E-hub project is to maximise the amount of renewable energy in a district by matching energy demand and supply, by shifting the demand of heat pumps, refrigerators or white good appliances. Excess renewable heat can be stored in distributed buffers, advanced Thermo-Chemical Materials (TCMs) or boreholes. An important element is the acceptance of such an advanced energy system by stakeholders such as DSOs (Distribution System Operators), BRPs (Balancing Responsible Parties) and end users. Therefore, developing new business models and service concepts is crucial.
MAIN OBJECTIVES
WP1 will aim at a first-level (conceptual) system definition of the E-Hub. At this stage, high-level requirements and overall system specifications need to be defined. In general terms, WP1 will define the overall framework in which the E-hub system will be integrated. The energy related performance indicators for evaluation purposes, for different stakeholders will be described. Moreover, data will be acquired concerning electric power and thermal energy requirements for the districts of the case studies in WP6.
The objective of WP2 is to decide, based on inputs from WP1 which devices are needed and how to operate them. Any missing technologies will be identified and it will be analysed if they can be adapted from other technologies, or if they need basic developments. Various criteria will be taken into account, such as cost, emissions, energy efficiency, availability, security, and other parameters like failure and repair rates that must be defined for all components in the system.
The overall objective of WP3 is to develop technologies and components that are crucial to realize an E-hub and that are not currently available. These components and techniques should bridge the gap between decentralized renewable energy production and consumption. In this project we focus on thermal storage which we consider a breakthrough technology, essential to reach the ultimate goal of energy neutral districts.
Another aim of WP3 is to develop alternative concepts to harvest renewable energy, i.e. integrated concepts like thermo-active foundations and thermal solar road collectors.
The objective of WP4 is to simulate districts in terms of energy management. To be able to do so, we need to develop (as much as possible) a technology independent active demand response system at district level. This requires the implementation of the requirements for an E-hub system as defined in WP1 into an energy system architecture, with specifications on ICT requirement and to integrate the business strategies defined in WP6 in the global control system.
The main objective of WP5 is the adaptation of research and simulation in a full scale demonstration in the district of Tweewaters in Leuven, Belgium. The aim is to implement an integrated and smart system to match supply and demand for multi commodities. The installation of a smart monitoring and data analysis system will provide short term energy information for producer and end user (through end user interface “MyJames”) and allows active management of the multi commodity network.
The objectives of WP6 are to provide the basic business information to define alternative energy service concepts and their business models. In a first step, we aim to identify market needs and existing and new business models (concept, content, earning logics, financing models and risk management) in the area of district level energy services. In a second step we aim to develop supporting methods and tools to improve the use of innovative procurement/sourcing models.
Another objective is to provide practical guidelines for different stakeholders (individual/households, public buildings’ owners, commercial/industrial buildings’ owners and project developers) to interact with investors to finance an energy efficient district
Finally, a specific task is included to contribute to the deployment and implementation of the results generated in the other work packages, by analysing the feasibility of different energy service /business models and energy production configurations in a number of case studies.
The main objective of WP7 is to make an impact with the project and its results by disseminating the knowledge gained. As the project covers a wide range of activities, from system analysis through component/ technology development and business model development to demonstration in a real situation, the specific objectives and instruments (website, workshops, scientific papers etc.) for dissemination may vary per activity.
The main objective of WP8 is to organize and manage the project and to communicate with the Commission.
Project Results:
WORK PACKAGE 1: SYSTEM DEFINITION
Work packages 1 and 2 can be characterized as preparatory for work in later work packages. In work package 1, we identified six Model District Types that can represent typical residential and non-residential districts across Europe. The model districts are intended to prepare for the evaluation of the case studies in WP6. Monthly energy demand for heating, cooling and electricity was calculated. In addition, a survey across eight European countries was made to study ownership and management of buildings and electricity grids and heating networks to be used in the business models in WP6. Finally, an evaluation methodology was made, yielding kpi’s on energy, ecology and economy, to be used in the evaluation of the case studies in WP6.
WORK PACKAGE2: ENERGY CONVERSION & STORAGE
In work package 2, we produced an inventory of existing technologies, to be used in the simulations in WP4. When using smart control of power and heat generating systems, they are expected to be operated differently from stationary systems running at nominal conditions, as is currently the case e.g. in large electricity plants. Therefore, intermittent operation, start-up behaviour and operation at partial load are important aspects to consider.
Therefore, representative real equipment such as a micro turbine CHP, an absorption cooler unit, an internal Combustion Engine (with a 1.2 litre Fiat engine), a fuel cell gas turbine hybrid system and a Stirling Engine were tested in the lab of TPG-DIME. Their performance was evaluated under different operating conditions. The characteristics of the equipment were implemented in the numerical models used to run the simulations in WP4.
WP2 also proposed, for each of the model districts, heat and electricity generating equipment, based on the energy profiles of the district and the analysis of a number of real cases with renewable energy and energy efficient equipment. These systems, based on experts’ assessments, are called the ‘best practice‘ systems.
Following the methodology developed in WP1, the kpi’s (key performance indicators) of energy, ecology and economy were calculated for a number of alternative systems. This resulted in an ‘optimised’ system, with a share of renewables of at least 20% of the total energy demand.
WORK PACKAGE 3: COMPONENTS AND TECHNIQUES DEVELOPMENT
In WP3 we continued development of a number of thermal storage components and produced a number of numerical models of thermal/electricity storage components and of heat /electricity generation equipment for use in the simulation environment in WP4.
Thermo-Active Foundations
The work on Thermo-Active Foundations (TAF) aimed to improve the heat transfer between soil and piping network by using alternative materials Simulations were carried out using calibrated high detail 3D-FEM models. Parameters studied include piping and concrete heat transfer, piping layout. The use of thermally enhanced piping material appears to have little effect on the thermal performance of the energy pile. However, thermally enhanced concrete (15% more expensive) increases the performance in base load operation by 10-25 % and in peak load operation by 25-30%
In the work on thermo-active foundations, partners SOL and HSW developed a simple implicit model to be used in the Matlab simulation environment developed in WP4, allowing the assessment of TAFs in a district energy system.
The simplified model was compared against analytical solutions and CFD (Computational Fluid Dynamics) data and it has also been compared with calibrated models made with EED commercial software.
The integration of the TAF system at building level was also studied. The goal was an optimal low-temperature piping network, with a single central heat pump or decentralised heat pumps.
Model district 4, located in Western Europe, was chosen to virtually build a low temperature district heating network. It contains five buildings with a ground area of 6,500 m² in a total area of 16,500 m². For this district, the two most efficient and practically proven possible ways of balancing thermal energy on a district level were analysed.
Thermal Road Solar Collector system
The Thermal Road Solar Collector (TRSC) developed in this task is to be used in combination with Thermo-Active Foundations to restore the heat balance of the soil over a summer and winter period. The work on the optimisations of the TRSC (structurally and thermally) aims to improve the heat transfer between pavement and piping network to maximise TRSC outlet temperature. The higher the temperature level, the higher the efficiency of the total energy system.
Simulations were made to determine efficiency curves for the road solar collector. Parameters studied include pipe diameter, depth and spacing, insulation layer under pipe array, asphalt IR reflectivity. In addition to the simulations, structural tests were carried out to investigate the strains and wear of TRSC lay-outs.Tests verified that the addition of a thin layer of resin in the top few mm of the asphalt allows the location of the pipe system at the bottom of the wearing course.
As a result, a validated model is available to calculate optimal piping layout, depending on weather and traffic conditions. In addition, a simple thermal model of the solar thermal road collector was made in Matlab, for use in the simulation tool in WP4. TRNSYS simulations were used to validate the simplified model
ThermoChemical Storage
ECN and TNO investigated long term thermal storage using Thermo Chemical Materials (TCM’s). The aim was to construct and test Thermo-Chemical Storage reactors, supported by a validated theoretical model.
ECN decided to concentrate on atmospheric TCM systems (using moist air) while TNO pursued the route of sub- atmospheric TCM systems (using water vapour without air). The first has the advantage of being more suited for practical purposes, but has the disadvantage of having to provide a forced airflow through the system, requiring auxiliary electrical energy. The second has the advantages of higher power density and fast charging and discharging of the store, while its main disadvantage is a higher complexity as a near-vacuum has to be maintained.
At ECN, a 15kWh heat storage system was designed, built and initial test were carried out. The system is based on an open sorption concept to achieve a simple and low cost solution for the storage containers. It contains 2 vessels of 112 dm3, each filled with 75 kg of zeolite 13X grains. The storage capacity reached 14 kWh and thermal powers for charging and discharging are in the range of 0.5-1 kW. The temperature inside the zeolite reactor was measured at various heights. . The temperature in the bed rises from bottom to top, following the air supply direction. The temperature gradient in the bed points to a moving reaction zone of desorption. In discharging mode, cool and humidified air is blown through the dried zeolite and the water vapour is adsorbed. The heat of sorption is released and over time, the zeolite and the air increase in temperature to values of up to 70°C. After 25 hours of discharge the zeolite has reached its equilibrium sorption capacity and heat release stops rather rapidly. The steep drop in temperature through the bed again indicates a distinct reaction zone inside the bed that gradually shifts from bottom to top.
In addition to the experimental work, a dynamic simulation model of an open sorption reactor was validated, which will be a valuable tool for future upscaling.
In conclusion, the open sorption concept allows simple storage reactor design, but further improvements are needed, including: 1) reduction of auxiliary electricity, 2) reduction of thermal losses in air handling, 3) application of materials with higher energy density.
At TNO, a 3 kWh-reactor filled with 40 kg of zeolite 5A spheres was built and tested. It contains eight heat exchangers connected in parallel.
Experiments showed stable sorption material behaviour with temperature lifts of 20 – 50°C, matching levels in building heating demands. The energy content appeared to depend on operating conditions, in particular those during the drying (desorption) process. The energy content measured was lower than that derived from Clausius-Clapeyron curves found in the literature. Most likely, the zeolite used differed in material characteristics from the zeolite on which the curves were based.
In addition to the experimental work, a numerical model was developed that describes the physical processes of evaporation, absorption, heat transfer etc. taking place in the evaporator unit and reactor vessel. Model calculations based on first guesses of parameters yielded a reasonable fit with experimental results. Finally, the expertise and experience gained was used to design a 15 kWh reactor, based on the 3 kWh module. The height of the vessel was maximised to 2m in order to allow installation in a dwelling. With 175 kg of zeolite at the current performance yields a value of 16 kWh for the energy content. Because of the higher packing density, the system energy density (based on the volume of the vessel rather than the volume of the zeolite) increased from 0.08 GJ/m3 for the 3kWh module to 0.13 GJ/m3 for the 16 kWh module.
Distributed Thermal Storage
VITO investigated the benefits of distributed thermal storage. Instead of a large centralised heat storage vessel, smaller individual vessels can be located in households or offices, with the advantage that the heat is stored close to where it is needed. Excess production of thermal energy (from solar collectors or from a CHP) need not be discarded but can be stored for future use.
VITO designed an experimental setup with 4 vessels including the emulation of heat production and consumption.
Different heat storage strategies for the vessels in a district heating grid fed by a CHP are compared for energetic and economic performance, using a ‘hardware-in-the-loop’ simulation model. The flexibility resulting from the storage vessels is used to actively control the CHP, which in this way can produce electricity at times of high electricity prices. The results of the simulations show that the control framework developed performs well, resulting in higher profits in operating the CHP.
The configuration with distributed heat storages performs best, however only slightly better than the active configuration without buffers (red dashed line). The results for the central buffer case (black dotted line) are a little worse, but still a lot better than in the reference case (blue line) without active control. The reasons for the lower performance is that the thermal mass of the buildings, which is activated in the first two configurations, is rather high compared to that of the buffers, resulting in much more flexibility and consequently higher profits. Summarizing, the results show that active control of the CHP is able to increase the profit of the CHP significantly.
VITO also studied a method to reliably determine the State of Charge (SoC) of a water based storage vessel with a minimum number of temperature sensors. A model was developed taking into account heat losses and heat exchange between water layers, heat convection when the system is drained, mixing and heating. The model was able to estimate the temperature profile in the buffers with a small number of sensors typically 4). It was also found that the positions of the sensors are crucial parameters to obtain good results.
Modelling of Components
When doing simulations of energy management systems on district level, we need numerical models of thermal/electricity storage components and of heat /electricity generation equipment as well as a dynamic numerical model of the building is a district. As the simulation environment in WP4 was going to be programmed in Matlab, the components models would also have to be modelled in Matlab.
The starting point for the components was the list of equipment proposed in WP2 for the model districts. For each model, a numerical model was written in Matlab code and validated by comparing the output to either measurements of real equipment, or results of commercial software (e.g. TRNSYS) or checked for correct operation.
In particular in residential districts, the buildings are the main consumers of thermal energy. To model the buildings in a district, a simple dynamic model was made that allows the aggregation of a large number of individual buildings of similar thermal performance. This way, the district can be represented by a limited (typically 5-15) number of aggregated buildings, allowing the definition of their space heating and space cooling demands during a simulation run. As long as the indoor temperature in the buildings remains between certain boundaries, the space heating demand and space cooling demand of each building has some degree of flexibility.
In addition the libraries of the ECoMP and TRANSEO software, which constitute proprietary knowledge of partner TPG-DIME, were enriched with a number of models of new components.
WORK PACKAGE 4: ENERGY MANAGEMENT
Simulation environment
This work package is the pivotal work package, where a simulation environment was developed. In an early stage of the project it was decided to program the simulation environment in Matlab as this programming language offered the required functionality and flexibility. In addition, Matlab expertise was available at all partners contributing to the development of the simulation environment. The latter includes:
• The implementation of the models of energy generation and storage equipment produced in WP3, represented by model agents. The agent of a component produces a bid functions which is the translation of the state of the component (e.g. for storage components: the State of Charge) into a willingness to consume or supply energy.
• The simplified model of aggregated buildings to calculate the space heating demand of the district.
• A heating network and electricity grid connecting energy producers and consumers. A Graphical User Interface (GUI) was made to facilitate the configuration of the district.
• Off-line calculation of electricity and Domestic Hot Water (DHW) demand profiles of the district. The electricity profile is in the form of a so-called ‘flex-graph’, giving the boundaries between which momentary electricity demand can vary as a result of time-shiftable appliances in the district such as white good appliances.
• Business agents representing Business models based on flexibility of demand:
1. time of use/ToU business agent and
2. peak shaving agent, which can also be considered as a ‘reduce imbalance cost’ business agent as they are based on similar principles.
• The MCM (Multi Commodity Matcher) control algorithm, discussed below. However, the simulation environment is open to work with alternative control algorithms if they comply with the interfaces described.
After the simulation, the performance of the energy management system can be assessed by analysing a number of predefined Key Performance Indicators (KPI’s) on energy use (kWh) economy (euros), ecology (CO2 emissions) and KPI’s related to peak shaving.
Control algorithm
In the project, an energy control algorithm called the Multi Commodity Matcher (MCM) was developed to match supply and demand of electricity and heat simultaneously on district level.
Several technologies are available for matching the supply and demand of energy. In this project we used agent based technology, used e.g. in the Powermatcher ® or IntelliGator ® software. In fact, the MCM developed in this project is an extension of the Powermatcher concept (which is for electrical power only) to electricity and heat, inheriting the Powermatcher’s advantages of scalability and user autonomy.
Instead of a one-dimensional bid curve for a single commodity (electricity) we need to define two 2-dimensional bid surfaces, one for electricity and one for heat. For each multi-commodity agent, the bid surfaces represent the willingness to consume or produce electricity and heat at a range of electricity and heat prices. As an illustration, the figures below show the bid surfaces of a heat pump.
TNO as holder of the IP to the MCM decided not to apply for a patent but instead intends to make this the standard in the field of agent based control systems. To this end, TNO is involved in the formation of an alliance called the Flexible Power Alliance Network, see http://www.flexiblepower.org/nl/
The MCM control algorithm was used in three types of applications which are discussed in the next chapters:
1. Simulations of a virtual application of an advanced Energy Management System in five case studies
2. Controlling the operation of several cogeneration units in a real lab environment, under conflicting demand profiles of heat and electricity
3. Full scale demonstration in the district of Tweewaters in Leuven, Belgium
APPLICATION OF SMART ENERGY MANAGEMENT IN CASE STUDIES
The simulations of the case studies were carried out according to three scenarios:
1. A BAU (Business As Usual) scenario, using a conventional energy supply system.
2. A Green or Low carbon scenario, implementing RES (Renewable Energy Sources) and REC (Recovered energy) technologies.
3. A Smart scenario, similar to scenario 2 but run with a smart energy management system, making use of the flexibility in demand.
Annual profiles with hourly time steps were produced.
For each case study different kpi’s were produced in order to compare the different scenarios, the most important kpi’s being:
- Electricity and heat demand.
- Electricity and heat generation.
- Electricity grid imports and exports.
- Electricity demand covered by the Grid, RES (Renewable Energy Sources).
- Heat demand covered by RES and REC.
- Use of locally generated energy.
- Primary Energy demand (PE) and associated CO2 emissions.
The introduction of technologies based on Renewable Energy Sources (RES) and/or Recovered Energy (REC) within the district (GREEN/Low Carbon scenario) results in an important decrease in primary energy use and CO2 emissions and leads to more beneficial cash-flows for the studied cases. Moreover, by introducing smart capabilities (SMART scenario) extra savings in costs can be realized. In the SMART scenario the environmental impact may decrease or increase depending on the business case selected.
In addition to the study of technical issues, an overview is made of the regulatory framework and non-technical barriers and practical guidelines for each of these case studies. Barriers identified include: administrative and legal barriers, economic, financial and market barriers and social barriers and acceptance issues that may hamper, in the short term, the application of the E-hub concepts in real life cases.
APPLICATION OF SMART ENERGY MANAGEMENT IN LAB DEMO
The second application of a smart energy management is the demonstration of the Multi Commodity Matcher in a real lab environment controlling the operation of several cogeneration units. This required producing an interface allowing the communication between the MCM control algorithm running in Matlab and the data acquisition and control system, running in LabVIEW. The equipment controlled included:
• An Internal Combustion Engine CHP (ICE), with thermal/electrical output of 40/20 kW.
• A micro Gas Turbine CHP (mGT) with a thermal/electrical output of 160/100 kW.
• A boiler with a thermal output of 30 kW.
• A 5 m3 thermal storage vessel.
The thermal load can be controlled independently for each piece of equipment using fan coolers. The electricity demand can be controlled with a controllable resistance bank. Excess of electricity is fed to the grid.
The lab demo tests were carried out with and without thermal storage. Both tests were carried out according to predefined heating and electricity demand . For a cogeneration unit, producing heat and electricity at a more or less constant ratio, these are conflicting profiles, for which the control strategy has to find an optimal solution.
The results of 4 cases were compared:
1. A simulation, using the ECoMP (Economic Cogeneration Modular Programme) software, calculating the economic optimal use of different cogeneration units. This is the reference case.
2. A real test in the lab with the Zack algorithm controlling the cogeneration units
3. A real test in the lab with the MCM controlling the cogeneration units
4. A simulation with the MCM controlling the cogeneration units to check the operation of the MCM
The graphs of the MCM test and MCM simulation are similar, showing that in a real lab environment, the MCM performs as could be expected. The MCM test and the Zack control test also show similar behaviour. Finally, the shape of the curve is similar to that of the economic optimum given by the reference ECoPM simulation.
As in the test without storage, the graphs of the MCM test and MCM simulation are similar, showing that in a real lab environment, the MCM performs as expected. Contrary to the previous test, the Zack control choses a different – non-optimal- strategy to control the test rig. This may be due to the fact that the Zack control was not designed to operate with thermal storage.
Finally, when comparing the MCM (both test and simulation) with ECoMP, the shape of the curves in the first half of the test are similar, showing the MCM comes close to the economic optimum. In the second half of the test, there is a difference between the MCM and ECoMP, due to differences between the actual thermal storage and the implementation of the model of thermal storage in ECoMP.
WORK PACKAGE 5: FULL SCALE DEMONSTRATION
Tweewaters is a unique inner-city development which is one of the largest inner-city developments in Belgium. It consists in total of 1,200 dwellings, commercial spaces, offices and other functions covering an area of 11ha in the city centre of Leuven. Ertzberg’s Urban Convenience® vision focuses on the integration and interaction of all aspects of a sustainable quarter (energy, mobility, use of open spaces, waste, consumption of food, etc.).
The district is still under development and the first building completed is the ‘Balk van Beel’ apartment building, housing 106 families and commercial spaces. A smart energy monitoring and control system was installed in this building.
The Balk van Beel was International recognised with the award of the BREEAM ‘Outstanding’ certificate and the 2013 BRE award. The quarter of Tweewaters and its first phase, the ‘Balk van Beel’, received a nomination for the Global Cleantech Cluster Association (GCCA) Later Stage Award and the European Corporate Social Responsibility Award (CSR). Both the building and the quarter are also being used as a model project by Leuven Climate Neutral 2030 and the Flanders in Action (VIA) program.
The energy concept of the Tweewaters district is based on local production and consumption of green heat and electricity, using a natural gas fired (later: biomass fired) CHP plant. Smart control will be applied to flexible energy sources to match energy supply and demand, to decrease the disturbance to the grid and to enhance the opportunity for green energy production. The integration with a neighbouring district heating network for the sale of heat is taken into account.
An energy consortium between Ertzberg (developer of the quarter and of the smart control of energy), Dalkia (the energy producer) and Eandis (the distribution system operator for heat and electricity) was set up to supply the district with energy.
Over the summer, Ertzberg conducted interviews and workshops with the tenants, in order to report on client behaviour. Privacy issues have been addressed and feedback from the tenants has been reported.
The smart monitoring / management system installed in 106 dwellings and 9 commercial spaces in the Balk van Beel, allowed us to produce Profiles of DHW (Domestic Hot Water), space heating and electricity.
The energy demand for space heating in the ‘Balk van Beel’ ranges from 6 to 26 kWh/m2a, with an average of 11 kWh/m2a. This is better than the Passive house standard of 15 kWh/m2a, showing the high quality of the building.
Regarding the potential of shifting the consumption of household appliances, the rather low amount of smart starts allowed by the tenants does not allow us to draw solid conclusions. In the specific case of Tweewaters, a cost reduction of 20% was achieved compared to the reference scenario. However the mean flexibility of 136 minutes observed in Tweewaters is too low to gain significant electricity cost reductions. For higher profits, the flexibility window should increase to 420 to 460 minutes.
In a simulation, with a smart control of the CHP, operating it at times of high electricity prices, a realistic estimation is that profits from electricity sold to the grid can be raised by 30%. Over a period of one year this results in a profit in the order of 47 € per apartment.
The savings achieved may in themselves not justify the investment in a smart energy system. However, smart control systems are a necessary part of future energy systems to optimally meet all end users’ ever increasing energy demands. In the next phases of the development of the Tweewaters quarter, the energy management system will be applied to the other buildings.
WORK PACKAGE 6: BUSINESS STRATEGIES and NON-TECHNICAL ISSUES
WP6 provides the basic business information to define alternative energy service concepts and their business concepts as well as implementing them in terms of ‘business agents’.
In the first phase of the project, an overview was made of existing business models from a literature study, 20 case studies by partners and a web questionnaire. The overview also included an overview of barriers and incentives for district energy concepts.
Taking the perspective from an investor to finance an energy efficient district, a number of “Supporting methods and tools for financing energy efficient districts” were identified, through a research activity aimed at:
• Classifying stakeholders according to their financing/investments needs and their role in providing financial resources.
• Understanding the basic features of investments in district heating networks and smart grids.
• Examining existing and emerging models to finance EE/RES projects at private and public level.
• Identifying and discussing 17 relevant best practices already implemented in this field.
• Analysing a set of real life financing case studies, according to criteria such as financing terms, equipment ownership, responsibilities/liabilities of each party, requested guarantees and type of assessment applied by the financier to evaluate the project.
• Discussing barriers and recommendations with relevant financial stakeholders.
Through an extensive research, conducted by analysing existing literature and conducting interviews with financiers, we outlined short practical guidelines for different stakeholders: individual/households, public buildings’ owners, commercial/industrial buildings’ owners and project developers.
A future district is imagined with different groups of energy consumers, producers and “prosumers” (members who are both producers and consumers). On district level, the match between the supply and demand of energy is managed by a “multi commodity matcher” using an automated pricing mechanism. The district is also connected to the national grid, so the national price of energy also affects the price of energy in the district.
The general pricing mechanisms are described. Currently, the marginal price of electricity is determined by demand and supply on national level. In times of a large supply from, for instance, an offshore wind farm, electricity prices on the exchange market will be low (and may even be negative) and in times of high demand, prices will be higher. Balance responsible parties (BRP’s) can trade the energy by day-ahead or by intraday market mechanisms. Currently consumers typically pay a flat tariff, but this is expected to change to TOU (time of use) pricing, critical peak pricing or even real-time tariffs.
Prosumers within the district are assumed to have a certain amount of flexibility in their consumption and production of energy. This flexibility is a new aspect in the energy market. A total of 15 novel business concepts were identified, some of them based on the concept of flexibility of energy demand. They are aimed at various stakeholders such as energy providers, balancing responsible parties (BRPs) transmission system operator (TSOs) and energy producers as well as business models for new roles. A limited number of these were produced in a ‘cook book’ or ‘recipe book’ of novel business models.
To apply the business concepts in our simulation environment, it needs to be translated into a “business agent”. It is also called an “objective agent” because it has the objective of steering a cluster of energy producers and consumers into a different mode, satisfying the demand of a certain stakeholder. For instance when a DSO or BRP needs a higher consumption of energy it may artificially decrease the energy price to promote consumption.
Price manipulation) is typically used in business cases that focus on time of use (TOU) pricing. The objective agent fixes a price on the cluster of devices that reflects the business cost for the cluster operator, e.g. it may streamline the power exchange price from an external market with a customer price to be paid in the cluster. A lower price will lead to higher demand (and lower supply) and vice versa.
Two business agents were produced:
1. time of use/ToU business agent and
2. peak shaving agent, which can also be considered as a ‘reduce imbalance cost’ business agent as they are based on similar principles.
Finally, a qualitative and a quantitative analysis of business models of the five case studies was carried out. The latter is based on the results of the simulations from WP4.
In the qualitative Business model, the roles of the different actors which are part of the energy consortium are given and the business cases which were applied within the different case studies are explained. The information is summarized in a graphical presentation of the business model, showing all actors which are part of the energy consortium and the most important interactions between them.
More information, including graphs, tables and pictures are reported in the file attached to this report.
Potential Impact:
ENERGY SUPPLY NOW AND IN THE FUTURE
Every day, we are extracting an enormous amount of oil, gas and coal from the earth. In that way, we are consuming solar energy that was captured over millions of years by plants and converted into ‘fossil fuels’.
The amount of fossil energy in our earth is huge but in the end it is limited. People are becoming increasingly aware that one day, fossil fuels will run out. Being an ‘energy-addicted’ society, we will have to find alternative sources, such as wind, the sun (Photo-Voltaic and solar thermal panels), biomass etc.
Mismatch between supply and demand of energy
The problem with renewable sources is that the energy they supply “is never there when you need it”. For instance, Photo-Voltaic panels deliver most of their electricity on sunny days around noon, while the highest demand for electricity (in dwellings) is in the evening when people turn on lights, television sets etc. A second example of a mismatch is the heat supplied by solar thermal panels in summertime and heat demand in wintertime.
How can we match energy supply and demand? We can do so by a combination of energy storage, bridging the time between supply and demand and by using a smart control of appliances in order to shift the time of demand.
Changes in the energy supply
Our society is changing in many ways. The number of electricity consuming appliances is growing fast, like smart phones, tablets and white good appliances. Numbers of large consumers like electric vehicles are growing faster and faster.In the future, the demand from all these users may be so high at certain times that one cannot charge one’s car because of limited capacity of the electricity grid. It will be the end of happy and unlimited energy consumption.
We expect that in the future, the harsh rules of capitalism - scarce commodities are more expensive - will apply to the energy supply. That means that energy will be more expensive in times of shortages of supply and cheaper in times of abundant supply. And so, future energy tariffs will vary from hour to hour rather than the flat tariff in use today.
When a home-owner wants to save on his energy bill, he can sit and wait next to his energy price indicator for the price to go down. Or he may use a smart control system to do it for him. The control system will know when the electric car needs to be charged (e.g. in the next day at 7 am) and it may use the weather-forecast, predicting when prices will be low due to abundant supply from off-shore wind farms. All this information will be used to ensure that the car is charged in the morning at the lowest possible price.
The Powermatcher © and MultiCommodity Matcher control software developed in this project are well prepared for these kind of changes. A smart citizen using smart controls will be prepared for the future.
HOW DID E-HUB CONTRIBUTE TO A CHANGING ENERGY SUPPLY?
In fact, thermal storage is sometimes called ‘the holy grail’ of energy neutral buildings. The consortium made good progress in improving several types of thermal storage, in particular on Thermo-Active foundations, Thermo-Chemical storage and distributed thermal storage .
Powermatcher ® and MultiCommodity Matcher software, the latter developed in the E-hub project, use a pricing mechanism - presently using artificial prices - to match the supply and demand of energy. These energy management systems are therefore well prepared for future price differentiation mechanisms.
TNO is setting up a network called the Flexible Alliance Network (http://www.flexiblepower.org/) aiming to set the Powermatcher ® control algorithm as the standard in control of smart grids.
An important element is the acceptance of such an advanced energy system by energy suppliers and users alike. Therefore, developing new business and service concepts that are attractive to all stakeholders is crucial. The consortium proposed a number of novel business concepts based on flexibility of demand.
A simulation tool was developed in the project to assess the application of energy storage and energy management systems in a number of ‘case studies’. In the frame of a ‘Joint Exploitation Agreement, the tool can be used by consortium members to provide consulting services e.g. to municipalities in configuring energy neutral/energy efficient districts.
Finally, the consortium set a best practice with the full scale demonstration of a high quality building equipped with a smart energy management system in the district of Tweewaters.
List of Websites:
In order to disseminate the results, a website was made (http://www.e-hub.org/) and regularly updated. It contains a public part as well as a restricted part, the latter serving as a database for all project related documents. For more information please contact F.G.H. Koene, frans.koene@tno.nl.
A glossy brochure was produced, giving an overview of the work in the e-hub project and its application to the full scale demonstration in Tweewaters.
The brochure can be requested by contacting partner Ertzberg: ive@ertzberg.be or info@ertzberg.be.
In the four years of the project, 6 peer reviewed publications were submitted to scientific journals and 23 papers were written and presented at international conferences.
More relevant information on dissemination can be found in the file attached to this report
Due to finite stocks of fossil fuels and the effect of greenhouse gas emissions, the amount of renewable energy from wind, biomass and solar energy etc. must strongly increase over present levels. However, due to the fluctuating nature of renewable energy supply, application of short term and long term energy storage and intelligent energy management systems are essential to match demand and supply of energy.
Thermal storage is sometimes called ‘the holy grail’ of energy neutral buildings. The consortium made good progress in improving several types of thermal storage, in particular on Thermo-Active foundations, Thermo-Chemical storage and distributed thermal storage.
The Multi Commodity Matcher (MCM) control algorithm developed in the E-hub project is able to match the supply and demand of electricity and heat simultaneously. The MCM control algorithm was used in three types of applications:
1. A simulation tool was developed in the project to carry out simulations of a virtual application of an advanced Energy Management System in five case studies: the districts of Amsterdam (NL), Freiburg (D), Bergamo (It), Leuven (B) and Dalian (China).
The main conclusion is that the MCM was able to accommodate the introduction of technologies based on Renewable Energy Sources (RES) and/or Recovered Energy (REC) within the district (GREEN/Low Carbon scenario). It resulted in an important decrease in primary energy use and CO2 emissions and leads to more beneficial cash-flows for the studied cases. Moreover, by introducing smart capabilities (in the “SMART” scenario) extra savings in costs can be realized. In the SMART scenario the environmental impact may decrease or increase depending on the business case selected.
2. The MCM was used to control the operation of several cogeneration units in a real lab environment, under conflicting demand profiles of heat and electricity.
The main conclusion is that both in tests with and without thermal storage, the MCM shows robustness in control, coming close to the best thermo-economic solution. The latter was found in a simulation with the “ECoMP” optimisation software.
3. An ‘electricity only’ version of the MCM was applied in a full scale demonstration in the district of Tweewaters in Leuven, Belgium.
The main conclusion is that the consortium set a best practice with the full scale demonstration of a high quality building equipped with a smart energy management system.
In the frame of a ‘Joint Exploitation Agreement, the simulation tool can be used by consortium members to continue to provide consulting services e.g. to municipalities in configuring energy neutral/energy efficient districts.
An important element is the acceptance of such an advanced energy system by stakeholders such as DSOs (Distribution System Operators), BRPs (Balancing Responsible Parties) and end users. Therefore, a number of new business models and service concepts, most of them based on the concept of ‘flexibility of demand’ were developed and applied in the case study simulations.
Project Context and Objectives:
INTRODUCTION
To achieve low energy or even energy neutral districts, the share of renewable energy must increase drastically over present levels. However, accommodating a large supplier of renewable energy, such as a wind farm, in the existing energy infrastructure is complicated by the fluctuating character of the energy supply with the result that renewable energy supply may be either too large or too small to cover the instantaneous demand. Both smart energy management systems and energy storage are essential to meet this challenge.
The objective of the E-hub project is to maximise the amount of renewable energy in a district by matching energy demand and supply, by shifting the demand of heat pumps, refrigerators or white good appliances. Excess renewable heat can be stored in distributed buffers, advanced Thermo-Chemical Materials (TCMs) or boreholes. An important element is the acceptance of such an advanced energy system by stakeholders such as DSOs (Distribution System Operators), BRPs (Balancing Responsible Parties) and end users. Therefore, developing new business models and service concepts is crucial.
MAIN OBJECTIVES
WP1 will aim at a first-level (conceptual) system definition of the E-Hub. At this stage, high-level requirements and overall system specifications need to be defined. In general terms, WP1 will define the overall framework in which the E-hub system will be integrated. The energy related performance indicators for evaluation purposes, for different stakeholders will be described. Moreover, data will be acquired concerning electric power and thermal energy requirements for the districts of the case studies in WP6.
The objective of WP2 is to decide, based on inputs from WP1 which devices are needed and how to operate them. Any missing technologies will be identified and it will be analysed if they can be adapted from other technologies, or if they need basic developments. Various criteria will be taken into account, such as cost, emissions, energy efficiency, availability, security, and other parameters like failure and repair rates that must be defined for all components in the system.
The overall objective of WP3 is to develop technologies and components that are crucial to realize an E-hub and that are not currently available. These components and techniques should bridge the gap between decentralized renewable energy production and consumption. In this project we focus on thermal storage which we consider a breakthrough technology, essential to reach the ultimate goal of energy neutral districts.
Another aim of WP3 is to develop alternative concepts to harvest renewable energy, i.e. integrated concepts like thermo-active foundations and thermal solar road collectors.
The objective of WP4 is to simulate districts in terms of energy management. To be able to do so, we need to develop (as much as possible) a technology independent active demand response system at district level. This requires the implementation of the requirements for an E-hub system as defined in WP1 into an energy system architecture, with specifications on ICT requirement and to integrate the business strategies defined in WP6 in the global control system.
The main objective of WP5 is the adaptation of research and simulation in a full scale demonstration in the district of Tweewaters in Leuven, Belgium. The aim is to implement an integrated and smart system to match supply and demand for multi commodities. The installation of a smart monitoring and data analysis system will provide short term energy information for producer and end user (through end user interface “MyJames”) and allows active management of the multi commodity network.
The objectives of WP6 are to provide the basic business information to define alternative energy service concepts and their business models. In a first step, we aim to identify market needs and existing and new business models (concept, content, earning logics, financing models and risk management) in the area of district level energy services. In a second step we aim to develop supporting methods and tools to improve the use of innovative procurement/sourcing models.
Another objective is to provide practical guidelines for different stakeholders (individual/households, public buildings’ owners, commercial/industrial buildings’ owners and project developers) to interact with investors to finance an energy efficient district
Finally, a specific task is included to contribute to the deployment and implementation of the results generated in the other work packages, by analysing the feasibility of different energy service /business models and energy production configurations in a number of case studies.
The main objective of WP7 is to make an impact with the project and its results by disseminating the knowledge gained. As the project covers a wide range of activities, from system analysis through component/ technology development and business model development to demonstration in a real situation, the specific objectives and instruments (website, workshops, scientific papers etc.) for dissemination may vary per activity.
The main objective of WP8 is to organize and manage the project and to communicate with the Commission.
Project Results:
WORK PACKAGE 1: SYSTEM DEFINITION
Work packages 1 and 2 can be characterized as preparatory for work in later work packages. In work package 1, we identified six Model District Types that can represent typical residential and non-residential districts across Europe. The model districts are intended to prepare for the evaluation of the case studies in WP6. Monthly energy demand for heating, cooling and electricity was calculated. In addition, a survey across eight European countries was made to study ownership and management of buildings and electricity grids and heating networks to be used in the business models in WP6. Finally, an evaluation methodology was made, yielding kpi’s on energy, ecology and economy, to be used in the evaluation of the case studies in WP6.
WORK PACKAGE2: ENERGY CONVERSION & STORAGE
In work package 2, we produced an inventory of existing technologies, to be used in the simulations in WP4. When using smart control of power and heat generating systems, they are expected to be operated differently from stationary systems running at nominal conditions, as is currently the case e.g. in large electricity plants. Therefore, intermittent operation, start-up behaviour and operation at partial load are important aspects to consider.
Therefore, representative real equipment such as a micro turbine CHP, an absorption cooler unit, an internal Combustion Engine (with a 1.2 litre Fiat engine), a fuel cell gas turbine hybrid system and a Stirling Engine were tested in the lab of TPG-DIME. Their performance was evaluated under different operating conditions. The characteristics of the equipment were implemented in the numerical models used to run the simulations in WP4.
WP2 also proposed, for each of the model districts, heat and electricity generating equipment, based on the energy profiles of the district and the analysis of a number of real cases with renewable energy and energy efficient equipment. These systems, based on experts’ assessments, are called the ‘best practice‘ systems.
Following the methodology developed in WP1, the kpi’s (key performance indicators) of energy, ecology and economy were calculated for a number of alternative systems. This resulted in an ‘optimised’ system, with a share of renewables of at least 20% of the total energy demand.
WORK PACKAGE 3: COMPONENTS AND TECHNIQUES DEVELOPMENT
In WP3 we continued development of a number of thermal storage components and produced a number of numerical models of thermal/electricity storage components and of heat /electricity generation equipment for use in the simulation environment in WP4.
Thermo-Active Foundations
The work on Thermo-Active Foundations (TAF) aimed to improve the heat transfer between soil and piping network by using alternative materials Simulations were carried out using calibrated high detail 3D-FEM models. Parameters studied include piping and concrete heat transfer, piping layout. The use of thermally enhanced piping material appears to have little effect on the thermal performance of the energy pile. However, thermally enhanced concrete (15% more expensive) increases the performance in base load operation by 10-25 % and in peak load operation by 25-30%
In the work on thermo-active foundations, partners SOL and HSW developed a simple implicit model to be used in the Matlab simulation environment developed in WP4, allowing the assessment of TAFs in a district energy system.
The simplified model was compared against analytical solutions and CFD (Computational Fluid Dynamics) data and it has also been compared with calibrated models made with EED commercial software.
The integration of the TAF system at building level was also studied. The goal was an optimal low-temperature piping network, with a single central heat pump or decentralised heat pumps.
Model district 4, located in Western Europe, was chosen to virtually build a low temperature district heating network. It contains five buildings with a ground area of 6,500 m² in a total area of 16,500 m². For this district, the two most efficient and practically proven possible ways of balancing thermal energy on a district level were analysed.
Thermal Road Solar Collector system
The Thermal Road Solar Collector (TRSC) developed in this task is to be used in combination with Thermo-Active Foundations to restore the heat balance of the soil over a summer and winter period. The work on the optimisations of the TRSC (structurally and thermally) aims to improve the heat transfer between pavement and piping network to maximise TRSC outlet temperature. The higher the temperature level, the higher the efficiency of the total energy system.
Simulations were made to determine efficiency curves for the road solar collector. Parameters studied include pipe diameter, depth and spacing, insulation layer under pipe array, asphalt IR reflectivity. In addition to the simulations, structural tests were carried out to investigate the strains and wear of TRSC lay-outs.Tests verified that the addition of a thin layer of resin in the top few mm of the asphalt allows the location of the pipe system at the bottom of the wearing course.
As a result, a validated model is available to calculate optimal piping layout, depending on weather and traffic conditions. In addition, a simple thermal model of the solar thermal road collector was made in Matlab, for use in the simulation tool in WP4. TRNSYS simulations were used to validate the simplified model
ThermoChemical Storage
ECN and TNO investigated long term thermal storage using Thermo Chemical Materials (TCM’s). The aim was to construct and test Thermo-Chemical Storage reactors, supported by a validated theoretical model.
ECN decided to concentrate on atmospheric TCM systems (using moist air) while TNO pursued the route of sub- atmospheric TCM systems (using water vapour without air). The first has the advantage of being more suited for practical purposes, but has the disadvantage of having to provide a forced airflow through the system, requiring auxiliary electrical energy. The second has the advantages of higher power density and fast charging and discharging of the store, while its main disadvantage is a higher complexity as a near-vacuum has to be maintained.
At ECN, a 15kWh heat storage system was designed, built and initial test were carried out. The system is based on an open sorption concept to achieve a simple and low cost solution for the storage containers. It contains 2 vessels of 112 dm3, each filled with 75 kg of zeolite 13X grains. The storage capacity reached 14 kWh and thermal powers for charging and discharging are in the range of 0.5-1 kW. The temperature inside the zeolite reactor was measured at various heights. . The temperature in the bed rises from bottom to top, following the air supply direction. The temperature gradient in the bed points to a moving reaction zone of desorption. In discharging mode, cool and humidified air is blown through the dried zeolite and the water vapour is adsorbed. The heat of sorption is released and over time, the zeolite and the air increase in temperature to values of up to 70°C. After 25 hours of discharge the zeolite has reached its equilibrium sorption capacity and heat release stops rather rapidly. The steep drop in temperature through the bed again indicates a distinct reaction zone inside the bed that gradually shifts from bottom to top.
In addition to the experimental work, a dynamic simulation model of an open sorption reactor was validated, which will be a valuable tool for future upscaling.
In conclusion, the open sorption concept allows simple storage reactor design, but further improvements are needed, including: 1) reduction of auxiliary electricity, 2) reduction of thermal losses in air handling, 3) application of materials with higher energy density.
At TNO, a 3 kWh-reactor filled with 40 kg of zeolite 5A spheres was built and tested. It contains eight heat exchangers connected in parallel.
Experiments showed stable sorption material behaviour with temperature lifts of 20 – 50°C, matching levels in building heating demands. The energy content appeared to depend on operating conditions, in particular those during the drying (desorption) process. The energy content measured was lower than that derived from Clausius-Clapeyron curves found in the literature. Most likely, the zeolite used differed in material characteristics from the zeolite on which the curves were based.
In addition to the experimental work, a numerical model was developed that describes the physical processes of evaporation, absorption, heat transfer etc. taking place in the evaporator unit and reactor vessel. Model calculations based on first guesses of parameters yielded a reasonable fit with experimental results. Finally, the expertise and experience gained was used to design a 15 kWh reactor, based on the 3 kWh module. The height of the vessel was maximised to 2m in order to allow installation in a dwelling. With 175 kg of zeolite at the current performance yields a value of 16 kWh for the energy content. Because of the higher packing density, the system energy density (based on the volume of the vessel rather than the volume of the zeolite) increased from 0.08 GJ/m3 for the 3kWh module to 0.13 GJ/m3 for the 16 kWh module.
Distributed Thermal Storage
VITO investigated the benefits of distributed thermal storage. Instead of a large centralised heat storage vessel, smaller individual vessels can be located in households or offices, with the advantage that the heat is stored close to where it is needed. Excess production of thermal energy (from solar collectors or from a CHP) need not be discarded but can be stored for future use.
VITO designed an experimental setup with 4 vessels including the emulation of heat production and consumption.
Different heat storage strategies for the vessels in a district heating grid fed by a CHP are compared for energetic and economic performance, using a ‘hardware-in-the-loop’ simulation model. The flexibility resulting from the storage vessels is used to actively control the CHP, which in this way can produce electricity at times of high electricity prices. The results of the simulations show that the control framework developed performs well, resulting in higher profits in operating the CHP.
The configuration with distributed heat storages performs best, however only slightly better than the active configuration without buffers (red dashed line). The results for the central buffer case (black dotted line) are a little worse, but still a lot better than in the reference case (blue line) without active control. The reasons for the lower performance is that the thermal mass of the buildings, which is activated in the first two configurations, is rather high compared to that of the buffers, resulting in much more flexibility and consequently higher profits. Summarizing, the results show that active control of the CHP is able to increase the profit of the CHP significantly.
VITO also studied a method to reliably determine the State of Charge (SoC) of a water based storage vessel with a minimum number of temperature sensors. A model was developed taking into account heat losses and heat exchange between water layers, heat convection when the system is drained, mixing and heating. The model was able to estimate the temperature profile in the buffers with a small number of sensors typically 4). It was also found that the positions of the sensors are crucial parameters to obtain good results.
Modelling of Components
When doing simulations of energy management systems on district level, we need numerical models of thermal/electricity storage components and of heat /electricity generation equipment as well as a dynamic numerical model of the building is a district. As the simulation environment in WP4 was going to be programmed in Matlab, the components models would also have to be modelled in Matlab.
The starting point for the components was the list of equipment proposed in WP2 for the model districts. For each model, a numerical model was written in Matlab code and validated by comparing the output to either measurements of real equipment, or results of commercial software (e.g. TRNSYS) or checked for correct operation.
In particular in residential districts, the buildings are the main consumers of thermal energy. To model the buildings in a district, a simple dynamic model was made that allows the aggregation of a large number of individual buildings of similar thermal performance. This way, the district can be represented by a limited (typically 5-15) number of aggregated buildings, allowing the definition of their space heating and space cooling demands during a simulation run. As long as the indoor temperature in the buildings remains between certain boundaries, the space heating demand and space cooling demand of each building has some degree of flexibility.
In addition the libraries of the ECoMP and TRANSEO software, which constitute proprietary knowledge of partner TPG-DIME, were enriched with a number of models of new components.
WORK PACKAGE 4: ENERGY MANAGEMENT
Simulation environment
This work package is the pivotal work package, where a simulation environment was developed. In an early stage of the project it was decided to program the simulation environment in Matlab as this programming language offered the required functionality and flexibility. In addition, Matlab expertise was available at all partners contributing to the development of the simulation environment. The latter includes:
• The implementation of the models of energy generation and storage equipment produced in WP3, represented by model agents. The agent of a component produces a bid functions which is the translation of the state of the component (e.g. for storage components: the State of Charge) into a willingness to consume or supply energy.
• The simplified model of aggregated buildings to calculate the space heating demand of the district.
• A heating network and electricity grid connecting energy producers and consumers. A Graphical User Interface (GUI) was made to facilitate the configuration of the district.
• Off-line calculation of electricity and Domestic Hot Water (DHW) demand profiles of the district. The electricity profile is in the form of a so-called ‘flex-graph’, giving the boundaries between which momentary electricity demand can vary as a result of time-shiftable appliances in the district such as white good appliances.
• Business agents representing Business models based on flexibility of demand:
1. time of use/ToU business agent and
2. peak shaving agent, which can also be considered as a ‘reduce imbalance cost’ business agent as they are based on similar principles.
• The MCM (Multi Commodity Matcher) control algorithm, discussed below. However, the simulation environment is open to work with alternative control algorithms if they comply with the interfaces described.
After the simulation, the performance of the energy management system can be assessed by analysing a number of predefined Key Performance Indicators (KPI’s) on energy use (kWh) economy (euros), ecology (CO2 emissions) and KPI’s related to peak shaving.
Control algorithm
In the project, an energy control algorithm called the Multi Commodity Matcher (MCM) was developed to match supply and demand of electricity and heat simultaneously on district level.
Several technologies are available for matching the supply and demand of energy. In this project we used agent based technology, used e.g. in the Powermatcher ® or IntelliGator ® software. In fact, the MCM developed in this project is an extension of the Powermatcher concept (which is for electrical power only) to electricity and heat, inheriting the Powermatcher’s advantages of scalability and user autonomy.
Instead of a one-dimensional bid curve for a single commodity (electricity) we need to define two 2-dimensional bid surfaces, one for electricity and one for heat. For each multi-commodity agent, the bid surfaces represent the willingness to consume or produce electricity and heat at a range of electricity and heat prices. As an illustration, the figures below show the bid surfaces of a heat pump.
TNO as holder of the IP to the MCM decided not to apply for a patent but instead intends to make this the standard in the field of agent based control systems. To this end, TNO is involved in the formation of an alliance called the Flexible Power Alliance Network, see http://www.flexiblepower.org/nl/
The MCM control algorithm was used in three types of applications which are discussed in the next chapters:
1. Simulations of a virtual application of an advanced Energy Management System in five case studies
2. Controlling the operation of several cogeneration units in a real lab environment, under conflicting demand profiles of heat and electricity
3. Full scale demonstration in the district of Tweewaters in Leuven, Belgium
APPLICATION OF SMART ENERGY MANAGEMENT IN CASE STUDIES
The simulations of the case studies were carried out according to three scenarios:
1. A BAU (Business As Usual) scenario, using a conventional energy supply system.
2. A Green or Low carbon scenario, implementing RES (Renewable Energy Sources) and REC (Recovered energy) technologies.
3. A Smart scenario, similar to scenario 2 but run with a smart energy management system, making use of the flexibility in demand.
Annual profiles with hourly time steps were produced.
For each case study different kpi’s were produced in order to compare the different scenarios, the most important kpi’s being:
- Electricity and heat demand.
- Electricity and heat generation.
- Electricity grid imports and exports.
- Electricity demand covered by the Grid, RES (Renewable Energy Sources).
- Heat demand covered by RES and REC.
- Use of locally generated energy.
- Primary Energy demand (PE) and associated CO2 emissions.
The introduction of technologies based on Renewable Energy Sources (RES) and/or Recovered Energy (REC) within the district (GREEN/Low Carbon scenario) results in an important decrease in primary energy use and CO2 emissions and leads to more beneficial cash-flows for the studied cases. Moreover, by introducing smart capabilities (SMART scenario) extra savings in costs can be realized. In the SMART scenario the environmental impact may decrease or increase depending on the business case selected.
In addition to the study of technical issues, an overview is made of the regulatory framework and non-technical barriers and practical guidelines for each of these case studies. Barriers identified include: administrative and legal barriers, economic, financial and market barriers and social barriers and acceptance issues that may hamper, in the short term, the application of the E-hub concepts in real life cases.
APPLICATION OF SMART ENERGY MANAGEMENT IN LAB DEMO
The second application of a smart energy management is the demonstration of the Multi Commodity Matcher in a real lab environment controlling the operation of several cogeneration units. This required producing an interface allowing the communication between the MCM control algorithm running in Matlab and the data acquisition and control system, running in LabVIEW. The equipment controlled included:
• An Internal Combustion Engine CHP (ICE), with thermal/electrical output of 40/20 kW.
• A micro Gas Turbine CHP (mGT) with a thermal/electrical output of 160/100 kW.
• A boiler with a thermal output of 30 kW.
• A 5 m3 thermal storage vessel.
The thermal load can be controlled independently for each piece of equipment using fan coolers. The electricity demand can be controlled with a controllable resistance bank. Excess of electricity is fed to the grid.
The lab demo tests were carried out with and without thermal storage. Both tests were carried out according to predefined heating and electricity demand . For a cogeneration unit, producing heat and electricity at a more or less constant ratio, these are conflicting profiles, for which the control strategy has to find an optimal solution.
The results of 4 cases were compared:
1. A simulation, using the ECoMP (Economic Cogeneration Modular Programme) software, calculating the economic optimal use of different cogeneration units. This is the reference case.
2. A real test in the lab with the Zack algorithm controlling the cogeneration units
3. A real test in the lab with the MCM controlling the cogeneration units
4. A simulation with the MCM controlling the cogeneration units to check the operation of the MCM
The graphs of the MCM test and MCM simulation are similar, showing that in a real lab environment, the MCM performs as could be expected. The MCM test and the Zack control test also show similar behaviour. Finally, the shape of the curve is similar to that of the economic optimum given by the reference ECoPM simulation.
As in the test without storage, the graphs of the MCM test and MCM simulation are similar, showing that in a real lab environment, the MCM performs as expected. Contrary to the previous test, the Zack control choses a different – non-optimal- strategy to control the test rig. This may be due to the fact that the Zack control was not designed to operate with thermal storage.
Finally, when comparing the MCM (both test and simulation) with ECoMP, the shape of the curves in the first half of the test are similar, showing the MCM comes close to the economic optimum. In the second half of the test, there is a difference between the MCM and ECoMP, due to differences between the actual thermal storage and the implementation of the model of thermal storage in ECoMP.
WORK PACKAGE 5: FULL SCALE DEMONSTRATION
Tweewaters is a unique inner-city development which is one of the largest inner-city developments in Belgium. It consists in total of 1,200 dwellings, commercial spaces, offices and other functions covering an area of 11ha in the city centre of Leuven. Ertzberg’s Urban Convenience® vision focuses on the integration and interaction of all aspects of a sustainable quarter (energy, mobility, use of open spaces, waste, consumption of food, etc.).
The district is still under development and the first building completed is the ‘Balk van Beel’ apartment building, housing 106 families and commercial spaces. A smart energy monitoring and control system was installed in this building.
The Balk van Beel was International recognised with the award of the BREEAM ‘Outstanding’ certificate and the 2013 BRE award. The quarter of Tweewaters and its first phase, the ‘Balk van Beel’, received a nomination for the Global Cleantech Cluster Association (GCCA) Later Stage Award and the European Corporate Social Responsibility Award (CSR). Both the building and the quarter are also being used as a model project by Leuven Climate Neutral 2030 and the Flanders in Action (VIA) program.
The energy concept of the Tweewaters district is based on local production and consumption of green heat and electricity, using a natural gas fired (later: biomass fired) CHP plant. Smart control will be applied to flexible energy sources to match energy supply and demand, to decrease the disturbance to the grid and to enhance the opportunity for green energy production. The integration with a neighbouring district heating network for the sale of heat is taken into account.
An energy consortium between Ertzberg (developer of the quarter and of the smart control of energy), Dalkia (the energy producer) and Eandis (the distribution system operator for heat and electricity) was set up to supply the district with energy.
Over the summer, Ertzberg conducted interviews and workshops with the tenants, in order to report on client behaviour. Privacy issues have been addressed and feedback from the tenants has been reported.
The smart monitoring / management system installed in 106 dwellings and 9 commercial spaces in the Balk van Beel, allowed us to produce Profiles of DHW (Domestic Hot Water), space heating and electricity.
The energy demand for space heating in the ‘Balk van Beel’ ranges from 6 to 26 kWh/m2a, with an average of 11 kWh/m2a. This is better than the Passive house standard of 15 kWh/m2a, showing the high quality of the building.
Regarding the potential of shifting the consumption of household appliances, the rather low amount of smart starts allowed by the tenants does not allow us to draw solid conclusions. In the specific case of Tweewaters, a cost reduction of 20% was achieved compared to the reference scenario. However the mean flexibility of 136 minutes observed in Tweewaters is too low to gain significant electricity cost reductions. For higher profits, the flexibility window should increase to 420 to 460 minutes.
In a simulation, with a smart control of the CHP, operating it at times of high electricity prices, a realistic estimation is that profits from electricity sold to the grid can be raised by 30%. Over a period of one year this results in a profit in the order of 47 € per apartment.
The savings achieved may in themselves not justify the investment in a smart energy system. However, smart control systems are a necessary part of future energy systems to optimally meet all end users’ ever increasing energy demands. In the next phases of the development of the Tweewaters quarter, the energy management system will be applied to the other buildings.
WORK PACKAGE 6: BUSINESS STRATEGIES and NON-TECHNICAL ISSUES
WP6 provides the basic business information to define alternative energy service concepts and their business concepts as well as implementing them in terms of ‘business agents’.
In the first phase of the project, an overview was made of existing business models from a literature study, 20 case studies by partners and a web questionnaire. The overview also included an overview of barriers and incentives for district energy concepts.
Taking the perspective from an investor to finance an energy efficient district, a number of “Supporting methods and tools for financing energy efficient districts” were identified, through a research activity aimed at:
• Classifying stakeholders according to their financing/investments needs and their role in providing financial resources.
• Understanding the basic features of investments in district heating networks and smart grids.
• Examining existing and emerging models to finance EE/RES projects at private and public level.
• Identifying and discussing 17 relevant best practices already implemented in this field.
• Analysing a set of real life financing case studies, according to criteria such as financing terms, equipment ownership, responsibilities/liabilities of each party, requested guarantees and type of assessment applied by the financier to evaluate the project.
• Discussing barriers and recommendations with relevant financial stakeholders.
Through an extensive research, conducted by analysing existing literature and conducting interviews with financiers, we outlined short practical guidelines for different stakeholders: individual/households, public buildings’ owners, commercial/industrial buildings’ owners and project developers.
A future district is imagined with different groups of energy consumers, producers and “prosumers” (members who are both producers and consumers). On district level, the match between the supply and demand of energy is managed by a “multi commodity matcher” using an automated pricing mechanism. The district is also connected to the national grid, so the national price of energy also affects the price of energy in the district.
The general pricing mechanisms are described. Currently, the marginal price of electricity is determined by demand and supply on national level. In times of a large supply from, for instance, an offshore wind farm, electricity prices on the exchange market will be low (and may even be negative) and in times of high demand, prices will be higher. Balance responsible parties (BRP’s) can trade the energy by day-ahead or by intraday market mechanisms. Currently consumers typically pay a flat tariff, but this is expected to change to TOU (time of use) pricing, critical peak pricing or even real-time tariffs.
Prosumers within the district are assumed to have a certain amount of flexibility in their consumption and production of energy. This flexibility is a new aspect in the energy market. A total of 15 novel business concepts were identified, some of them based on the concept of flexibility of energy demand. They are aimed at various stakeholders such as energy providers, balancing responsible parties (BRPs) transmission system operator (TSOs) and energy producers as well as business models for new roles. A limited number of these were produced in a ‘cook book’ or ‘recipe book’ of novel business models.
To apply the business concepts in our simulation environment, it needs to be translated into a “business agent”. It is also called an “objective agent” because it has the objective of steering a cluster of energy producers and consumers into a different mode, satisfying the demand of a certain stakeholder. For instance when a DSO or BRP needs a higher consumption of energy it may artificially decrease the energy price to promote consumption.
Price manipulation) is typically used in business cases that focus on time of use (TOU) pricing. The objective agent fixes a price on the cluster of devices that reflects the business cost for the cluster operator, e.g. it may streamline the power exchange price from an external market with a customer price to be paid in the cluster. A lower price will lead to higher demand (and lower supply) and vice versa.
Two business agents were produced:
1. time of use/ToU business agent and
2. peak shaving agent, which can also be considered as a ‘reduce imbalance cost’ business agent as they are based on similar principles.
Finally, a qualitative and a quantitative analysis of business models of the five case studies was carried out. The latter is based on the results of the simulations from WP4.
In the qualitative Business model, the roles of the different actors which are part of the energy consortium are given and the business cases which were applied within the different case studies are explained. The information is summarized in a graphical presentation of the business model, showing all actors which are part of the energy consortium and the most important interactions between them.
More information, including graphs, tables and pictures are reported in the file attached to this report.
Potential Impact:
ENERGY SUPPLY NOW AND IN THE FUTURE
Every day, we are extracting an enormous amount of oil, gas and coal from the earth. In that way, we are consuming solar energy that was captured over millions of years by plants and converted into ‘fossil fuels’.
The amount of fossil energy in our earth is huge but in the end it is limited. People are becoming increasingly aware that one day, fossil fuels will run out. Being an ‘energy-addicted’ society, we will have to find alternative sources, such as wind, the sun (Photo-Voltaic and solar thermal panels), biomass etc.
Mismatch between supply and demand of energy
The problem with renewable sources is that the energy they supply “is never there when you need it”. For instance, Photo-Voltaic panels deliver most of their electricity on sunny days around noon, while the highest demand for electricity (in dwellings) is in the evening when people turn on lights, television sets etc. A second example of a mismatch is the heat supplied by solar thermal panels in summertime and heat demand in wintertime.
How can we match energy supply and demand? We can do so by a combination of energy storage, bridging the time between supply and demand and by using a smart control of appliances in order to shift the time of demand.
Changes in the energy supply
Our society is changing in many ways. The number of electricity consuming appliances is growing fast, like smart phones, tablets and white good appliances. Numbers of large consumers like electric vehicles are growing faster and faster.In the future, the demand from all these users may be so high at certain times that one cannot charge one’s car because of limited capacity of the electricity grid. It will be the end of happy and unlimited energy consumption.
We expect that in the future, the harsh rules of capitalism - scarce commodities are more expensive - will apply to the energy supply. That means that energy will be more expensive in times of shortages of supply and cheaper in times of abundant supply. And so, future energy tariffs will vary from hour to hour rather than the flat tariff in use today.
When a home-owner wants to save on his energy bill, he can sit and wait next to his energy price indicator for the price to go down. Or he may use a smart control system to do it for him. The control system will know when the electric car needs to be charged (e.g. in the next day at 7 am) and it may use the weather-forecast, predicting when prices will be low due to abundant supply from off-shore wind farms. All this information will be used to ensure that the car is charged in the morning at the lowest possible price.
The Powermatcher © and MultiCommodity Matcher control software developed in this project are well prepared for these kind of changes. A smart citizen using smart controls will be prepared for the future.
HOW DID E-HUB CONTRIBUTE TO A CHANGING ENERGY SUPPLY?
In fact, thermal storage is sometimes called ‘the holy grail’ of energy neutral buildings. The consortium made good progress in improving several types of thermal storage, in particular on Thermo-Active foundations, Thermo-Chemical storage and distributed thermal storage .
Powermatcher ® and MultiCommodity Matcher software, the latter developed in the E-hub project, use a pricing mechanism - presently using artificial prices - to match the supply and demand of energy. These energy management systems are therefore well prepared for future price differentiation mechanisms.
TNO is setting up a network called the Flexible Alliance Network (http://www.flexiblepower.org/) aiming to set the Powermatcher ® control algorithm as the standard in control of smart grids.
An important element is the acceptance of such an advanced energy system by energy suppliers and users alike. Therefore, developing new business and service concepts that are attractive to all stakeholders is crucial. The consortium proposed a number of novel business concepts based on flexibility of demand.
A simulation tool was developed in the project to assess the application of energy storage and energy management systems in a number of ‘case studies’. In the frame of a ‘Joint Exploitation Agreement, the tool can be used by consortium members to provide consulting services e.g. to municipalities in configuring energy neutral/energy efficient districts.
Finally, the consortium set a best practice with the full scale demonstration of a high quality building equipped with a smart energy management system in the district of Tweewaters.
List of Websites:
In order to disseminate the results, a website was made (http://www.e-hub.org/) and regularly updated. It contains a public part as well as a restricted part, the latter serving as a database for all project related documents. For more information please contact F.G.H. Koene, frans.koene@tno.nl.
A glossy brochure was produced, giving an overview of the work in the e-hub project and its application to the full scale demonstration in Tweewaters.
The brochure can be requested by contacting partner Ertzberg: ive@ertzberg.be or info@ertzberg.be.
In the four years of the project, 6 peer reviewed publications were submitted to scientific journals and 23 papers were written and presented at international conferences.
More relevant information on dissemination can be found in the file attached to this report
