D3.5: Description and report of energy exchange and smart trading tool including simulation studies for living labs as active LECs. Preliminary version

D3.5 Description and report of energy exchange and smart trading tool including simulation studies for living labs as active LECs. Preliminary version © Grid Singularity Pte Ltd HYPERGRYD. This project has received funding from the European Union’s Horizon 2020 research and innovation programme under grant agreement No 101036656 Ref. Ares(2024)4929207 - 08/07/2024 DISCLAIMER The opinion stated in this report reflects the opinion of the authors and not the opinion of the European Commission. All intellectual property rights are owned by HYPERGRYD consortium members and are protected by the applicable laws. Reproduction is not authorised without prior written agreement. The commercial use of any information contained in this document may require a licence from the owner of that information. Grid Singularity software tools described in this report are protected by open source GPL v.3 licence (backend code) and copyright (frontend interfaces). ACKNOWLEDGEMENT This project has received funding from the European Union’s Horizon 2020 research and innovation programme under grant agreement Nº 101036656. D3.5 Description and report of energy exchange and smart trading tool including simulation studies for living labs as active LECs. Preliminary version 2 Project Project Acronym HYPERGRYD Project Title Hybrid coupled networks for thermal-electric integrated Smart Energy Districts Grant Agreement number 101036656 Call identifier H2020-LC-GD-2020 Topic identifier LC-GD-2-1-2020 Innovative land-based and offshore renewable energy technologies and their integration into the energy system Funding Scheme Research and Innovation Action Project duration 42 months (From 1 October 2021) Coordinator ARCbcn Website http://hypergryd.eu Deliverable Deliverable No. 3.5 Deliverable title Description and report of energy exchange and smart trading tool including simulation studies for living labs as active LECs. Preliminary version Description In the framework of HYPERGRYD WP3, T3.4, GSY, in collaboration with project partners, has enhanced the Grid Singularity Exchange software (Energy exchange and smart trading tool for LEC with coupled networks) by developing, integrating and testing digital twins of the heat pump and district heating (virtual heat pump and heat demand digital twin). The solution was then demonstrated under T5.5 Model demonstration and validation of the HYPERGRYD ICT services at SONNE by simulating multiple scenaria and deploying the developed digital twins as part of the enhanced GSY Simulation Tool (also termed Singularity Map) to evaluate the role of district heating in local energy markets, the impact of flexibility assets and otherwise assess sector coupling and peer-to-peer benefits for individuals organised in energy communities. This report presents the related development, the testing and the demonstration study results. WP No. WP3 - ICT Modules and Simulation Tools Related task T3.4 – Energy exchange and smart trading tool for LEC with coupled networks especially Task 5.5 Model demonstration and validation of the HYPERGRYD ICT services at SONNE. Lead Beneficiary 1 - GSY Author(s) GSY (Fatuma Ali-Will, Hannes Diedrich, Ana Trbovich and Spyridon Tzavikas) Contributor(s) KTH, SONNE, AIT D3.5 Description and report of energy exchange and smart trading tool including simulation studies for living labs as active LECs. Preliminary version 3 Type R Dissemination PU Public Language English – GB Due 31/03/2024 Submission date 03/04/2024 Version Date Authors Description V.0.1 24/11/2023 GSY Version for internal debate and revision at WP level V.1.0 1/03/2024 GSY Version for internal review (to be approved by technical coordination) V.1.1 25/03/2024 SONNE, ARCbcn, AIT Internal review V.2.0 2/04/2024 GSY Final deliverable D3.5 Description and report of energy exchange and smart trading tool including simulation studies for living labs as active LECs. Preliminary version 4 Table of Contents 1 Executive Summary 12 2 Introduction 13 2.1 Scope 13 2.2 Audience 15 2.3 Definitions / Glossary 15 2.4 Abbreviations 16 2.5 Partner Contribution 17 2.6 Baseline 17 2.7 Relation to other activities 17 2.8 Structure 18 3 Heat Pump, Virtual Heat Pump and Heat Demand Digital Twin Development and Integration in the Grid Singularity Exchange 19 3.1. Heat Pump Digital Twin Development 19 3.2. District Heating (Virtual Heat Pump and Heat Demand) Digital Twin Development 24 3.2.1 Virtual Heat Pump and Heat Demand Digital Twin Modelling 24 3.2.2. Virtual Heat Pump Model Equations 25 3.2.3. Virtual Heat Pump Code Configuration 28 4 Heat Pump Development Testing via Grid Singularity Exchange Simulation 29 4.1. Heat Pump Digital Twin Testing Methodology 29 4.1.1. Validation Testing Scope and Key Performance Indicators 29 4.1.2. Validation Data Requirements, Collection and Mitigation Measures 30 4.1.3 Heat Pump Testing Simulation Configuration 31 4.2. Heat Pump Digital Twin Test Results 32 4.2.1. Heat Pump Digital Twin Measurement Test Results 32 4.2.2. Heat Pump Digital Twin Trading Test Results 37 4.2.3. Heat Pump Digital Twin Testing Conclusions 45 5 Grid Singularity Simulation Tool Demonstration: Study of Thermal Asset Coupling and Flexibility in a Sonnenplatz Local Electricity Market 47 5.1. Demonstration Study Context and Methodology 47 5.1.1. Demonstration Study Context 47 5.1.2. Demonstration Study Energy Asset Data 48 5.1.3. Demonstration Study Market Mechanism and Pricing Data 52 5.1.4 Demonstration Study Simulation Scenarios 54 5.2 Demonstration Study Results Presentation and Analysis 56 5.2.1. Base case scenario (SONNE LEC v.1) 56 5.2.2. Altered asset configuration (district heating replaced by heat pumps), no P2P trading (SONNE LEC v.2) 61 5.2.3. Base case with activated P2P trading (SONNE LEC v.3) 65 5.2.4. Activated P2P trading, with heat pumps replacing district heating (SONNE LEC v.4) 69 D3.5 Description and report of energy exchange and smart trading tool including simulation studies for living labs as active LECs. Preliminary version 5 5.2.5: Base Case with activated P2P trading and applied dynamic tariffs (SONNE LEC v.5) 76 5.2.6. Activated P2P trading with applied dynamic tariffs and heat pumps replacing district heating (SONNE LEC v.6) 85 5.2.7. Activated P2P trading with heat pumps replacing district heating and trading with an optimised strategy (SONNE LEC v.7) 93 6 Conclusions 98 References 100 D3.5 Description and report of energy exchange and smart trading tool including simulation studies for living labs as active LECs. Preliminary version 6 List of Figures Figure 1. Grid Singularity Exchange Heat Pump Digital Twin: Initial, Final and Preferred Buying Rate in relation to Minimum and Maximum Energy Consumed 20 Figure 2. Grid Singularity Exchange heat pump digital twin example of function to maintain temperature within comfort limits, with the horizontal axis showing timestamps for the selected date and the vertical axis showing temperature in °C. 20 Figure 3. Heat Pump Advanced Configuration Options in the Grid Singularity Exchange UI 24 Figure 4 : Ground source virtual heat pump diagram 25 Figure 5. Grid Singularity configuration of a digital twin of a partial LEM comprising of SONNE Homes 5, 9 and 10, with diverse energy asset type and number 31 Figure 6. SONNE Home 5 asset overview in the Grid Singularity simulation interface 32 Figure 7. SONNE Home 5 heat pump configuration in the Grid Singularity simulation interface 33 Figure 8. SONNE Home 9 asset overview in the Grid Singularity simulation interface 33 Figure 9. SONNE Home 9 heat pump configuration in the Grid Singularity simulation interface 34 Figure 10. SONNE Home 10 asset overview in the Grid Singularity simulation interface 34 Figure 11: SONNE Home 10 heat pump configuration In the Grid Singularity simulation interface 35 Figure 12a. SONNE Home 5 energy consumption profile on 1-7 July 2023 plotted from original raw data with the horizontal axis showing timestamps for the selected date and the vertical axis showing consumed energy in kWh 36 Figure 12b. SONNE Home 5 energy trading profile on 1-7 July 2023 extracted from the Grid Singularity simulation tool with the horizontal axis showing timestamps for the selected date and the vertical axis showing energy bought in kWh 36 Figure 13a. SONNE Home 10 PV production profile on 1-7 July 2023 plotted from original raw data with the horizontal axis showing timestamps for the selected date and the vertical axis showing produced energy in kWh 36 Figure 13b. SONNE Home 10 PV trading profile on 1-7 July 2023 plotted from original raw data extracted from the Grid Singularity simulation tool with the horizontal axis showing timestamps and the vertical axis showing energy sold in kW 37 Figure 14. Coefficient of performance as a function of time for heat pumps in simulated SONNE LEM for the period 1-7 July 2023 with the horizontal axis showing timestamps for the selected week and the vertical axis showing COP factor 37 Figure 15. Results in the Grid Singularity simulation interface showing community energy traded and bills table and the self-sufficiency and self-consumption rates 38 Figure 16. SONNE Home 5 energy bills and the self-sufficiency and self-consumption rates In the Grid Singularity simulation interface 39 Figure 17. SONNE Home 5 heat pump energy trading profile in the Grid Singularity simulation interface with the horizontal axis showing timestamps for the selected week and the vertical axis showing energy bought in kWh 40 Figure 18: SONNE Home 5 heat pump storage tank temperature time series in the Grid Singularity simulation interface with the horizontal axis showing timestamps for the selected week and the vertical axis showing the tank temperature in °C 41 Figure 19: SONNE Home 9 energy bills and the self-sufficiency and self-consumption rates In the Grid Singularity simulation interface 42 D3.5 Description and report of energy exchange and smart trading tool including simulation studies for living labs as active LECs. Preliminary version 7 Figure 20a SONNE Home 9 heat pump energy trading profile in the Grid Singularity simulation interface with the horizontal axis showing timestamps for the selected week and the vertical axis showing energy bought in kWh 42 Figure 20b SONNE Home 9 heat pump storage tank temperature time series in the Grid Singularity simulation interface with the horizontal axis showing timestamps for the selected week and the vertical axis showing the tank temperature in °C 43 Figure 21a SONNE Home 10 energy bills and the self-sufficiency and self-consumption rates In the Grid Singularity simulation interface 44 Figure 21b. SONNE Home 10 cumulative trade graph in the Grid Singularity simulation interface 44 Figure 22a. SONNE Home 10 heat pump energy trading profile in the Grid Singularity simulation interface with the horizontal axis showing timestamps for the selected week and the vertical axis showing the energy bought in kWh 45 Figure 22b. SONNE Home 10 heat pump storage tank temperature time series in the Grid Singularity simulation interface with the horizontal axis showing timestamps for the selected week and the vertical axis showing the tank temperature in °C 45 Figure 23. Sonnenplatz Großschönau GmbH digital twin configuration showing number and type of LEC participants and energy assets 48 Figure 24. Load consumption profile of SONNE Building 1, in Wh for 1-7 August 2023 with the horizontal axis showing timestamps for the selected week and the vertical axis showing the energy consumed in Wh 49 Figure 25. Total PV generation in the SONNE in kW for 1-7 August 2023, with the horizontal axis showing timestamps for the selected week and the vertical axis showing the power in kW 49 Figure 26. Combined daily consumption profile in kWh of 3 SONNE heat pumps on 1 August 2023 with the horizontal axis showing timestamps per 15-minute slot for the selected day and the vertical axis showing the energy consumption in kWh 51 Figure 27a. Example of Day Time-of-Use Tariff applied in the SONNE LEC simulation with the horizontal axis showing timestamps for the selected day and the vertical axis showing the price €cents/kWh 53 Figure 27b. Example of Night Time-of-Use Tariff applied in the SONNE LEC simulation with the horizontal axis showing timestamps for the selected day and the vertical axis showing the price €cents/kWh 53 Figure 28. Segment of the Grid Singularity Exchange configuration setup file for the simulation scenario v.1 54 Figure 29a. Net energy import and export graph for the SONNE LEC, simulation v1 for 1-7 August 2023 with the horizontal axis showing the building IDs and the vertical axis showing exported and imported energy in kWh 60 Figure 29b. Net energy import and export graph for the SONNE LEC, simulation v.1 for 1-7 November 2023 with the horizontal axis showing the building IDs and the vertical axis showing exported and imported energy in kWh 60 Figure 30a. Net energy import and export graph for the SONNE LEC, simulation v2 for 1-7 August 2023 with the horizontal axis showing the building IDs and the vertical axis showing exported and imported energy in kWh 64 Figure 30b. Net energy import and export graph for the SONNE LEC, simulation v.2 for 1-7 November 2023 with the horizontal axis showing the building IDs and the vertical axis showing exported and imported energy in kWh 65 D3.5 Description and report of energy exchange and smart trading tool including simulation studies for living labs as active LECs. Preliminary version 8 Figure 31a. Net energy import and export graph for the SONNE LEC, simulation v.3 for 1-7 August 2023 with the horizontal axis showing the building IDs and the vertical axis showing exported and imported energy in kWh 69 Figure 31b. Net energy import and export graph for the SONNE LEC, simulation v.3 for 1-7 November 2023 with the horizontal axis showing the building IDs and the vertical axis showing exported and imported energy in kWh 69 Figure 32. Battery state-of-charge (green), traded energy in kWh (purple) and energy rate in Euro/kWh (blue), Commercial Building 4 in SONNE LEC, simulation v.4 for 1-7 August 2023 with the horizontal axis showing timestamps for the selected week and the vertical axis showing the aforementioned variables 71 Figure 33. Battery state-of-charge (green), traded energy in kWh (purple) and energy rate in Euro/kWh (blue), Commercial Building 4 in SONNE LEC, simulation v.3 for 1-7 August 2023 with the horizontal axis showing timestamps for the selected week and the vertical axis showing the aforementioned variables 72 Figure 34a. Net energy import and export graph for the SONNE LEC, simulation v.4 for 1-7 August 2023 with the horizontal axis showing the building IDs and the vertical axis showing exported and imported energy in kWh 75 Figure 34b. Net energy import and export graph for the SONNE LEC, simulation v.4 for 1-7 November 2023 with the horizontal axis showing the building IDs and the vertical axis showing the exported and imported energy in kWh 76 Figure 35. Battery state-of-charge (green), traded energy in kWh (purple) and energy rate in Euro/kWh (blue), Commercial Building 4 in SONNE LEC, simulation v.5 for 1-7 August 2023 with the horizontal axis showing timestamps for the selected week and the vertical axis showing the aforementioned variables 83 Figure 36a. Net energy import and export graph for the SONNE LEC, simulation v.5 with Day ToU tariffs, for 1-7 August 2023 with the horizontal axis showing the building IDs and the vertical axis showing the exported and imported energy in kWh 84 Figure 36b. Net energy import and export graph for the SONNE LEC, simulation v.5 with Day ToU tariffs for 1-7 November 2023 with the horizontal axis showing the building IDs and the vertical axis showing exported and imported energy in kWh 84 Figure 36c. Net energy import and export graph for the SONNE LEC, simulation v.5 with Night ToU tariffs for 1-7 August 2023 with the horizontal axis showing the building IDs and the vertical axis showing exported and imported energy in kWh 85 Figure 36d. Net energy import and export graph for the SONNE LEC, simulation v.5 with Night ToU tariffs, for 1-7 November 2023 with the horizontal axis showing the building IDs and the vertical axis showing exported and imported energy in kWh 85 Figure 37a. Net energy import and export graph for the SONNE LEC, simulation v.6 with Day ToU tariff, for 1-7 August 2023 with the horizontal axis showing the building IDs and the vertical axis showing exported and imported energy in kWh 91 Figure 37b. Net energy import and export graph for the SONNE LEC, simulation v.6 with Day ToU tariffs, for 1-7 November 2023 with the horizontal axis showing the building IDs and the vertical axis showing exported and imported energy in kWh 92 Figure 37c. Net energy import and export graph for the SONNE LEC, simulation v.6 with Night ToU tariff for 1-7 August 2023 with the horizontal axis showing the building IDs and the vertical axis showing exported and imported energy in kWh 92 D3.5 Description and report of energy exchange and smart trading tool including simulation studies for living labs as active LECs. Preliminary version 9 Self-sufficiency rate refers to the share (expressed in percentage terms) of an individual local energy market (LEM) participant's (or community’s) energy demand produced (and self-consumed) locally calculated over a defined time frame (weekly or monthly). The self-sufficiency rate in the framework of this task and deliverable is calculated as follows: self_sufficiency(market) = self_consumed_energy / total_energy_demanded The energy cost savings performance indicator shows the financial savings from participating in a local energy market (LEM) over a defined time frame (weekly or monthly). The energy cost savings value in the framework of this task and deliverable is calculated as follows: Energy cost savings (€) = Energy bill prior to local trading (total energy demand supplied by external supplier (kWh) * (energy price set by external supplier + applicable grid fees) €- energy bill) with activated local energy trading (€) (LEM Bill). The net energy traded (energy bought minus energy sold) in the local energy market (LEM) performance indicator is calculated over a defined time frame (weekly or monthly) and expressed in kilowatt hours (kWh) for each participant and the LEM (energy community), respectively. 2.4 Abbreviations AIT: Austrian Institute of Technology CNR: Istituto di tecnologie avanzate per l'energia "Nicola Giordano" COP: Coefficient of Performance DER: Distributed Energy Resources DH: District Heating DHC: District Heating and Cooling ER: Expected Result GDPR: General Data Protection Regulation GSY: Grid Singularity (Grid Singularity GmbH, Grid Singularity Pte Ltd) HP: Heat Pump HYPERGRYD: Hybrid Coupled Networks for Thermal-Electric Integrated Smart Energy Districts ICT: Information and Communications Technology KTH: KTH Royal Institute of Technology in Stockholm LEC: Local Energy Community LEM: Local Energy Market LIL: Living Lab / Live-In Lab MQTT: Message Queuing Telemetry Transport P2P: Peer-to-Peer PV: Photovoltaics RES: Renewable Energy Source SOC: State of Charge SONNE: Sonnenplatz Großschönau GmbH, Austria T: Task UI: User Interface VHP: Virtual Heat Pump WP: Work Package D3.5 Description and report of energy exchange and smart trading tool including simulation studies for living labs as active LECs. Preliminary version 16 2.5 Partner Contribution The Sonnenplatz Großschönau GmbH (SONNE), provided the data and the related information used in validation testing and in the demonstration study of the digital twins of the heat pump and the district heating system (virtual heat pump and heat demand) development and integration in the Grid Singularity Exchange. The data collection was in part supported by the KTH Royal Institute of Technology in Stockholm (KTH), which developed a tool to derive data from the SONNE energy metres and operation sensors via the Internet by means of Message Queuing Telemetry Transport (MQTT) protocol. KTH also provided equations, which GSY then further developed and coded as part of the virtual heat pump modelling. Finally, CNR participated in research discussions and provided several academic references that were used in development conducted as part of Task 3.4 which is the basis of this deliverable. 2.6 Baseline This deliverable is based on previous work of GSY to develop the Grid Singularity Exchange as an open-source backend code with protected user interfaces [2], including, inter alia, the Singularity Map as a simulation tool (GSY Simulation Tool) [3]. The Grid Singularity Exchange has been enhanced in the framework of the EU HYPERGRYD Project [1], by developing a digital twin of the heat pump and a digital twin of the district heating system, including the digital twin of heat demand and a digital twin of a virtual heat pump that mimics and replaces the heat demand currently satisfied by a district heating network. The development also draws on relevant academic research for this digital twin modelling, which is fully referenced. Related EU projects such as the EU FEDECOM Project [7], supported by the European Union’s Horizon Europe programme under Grant Agreement No. 101075660, will complement and build upon this work, enabling GSY as a project partner active in both projects to develop a decentralised implementation of its software, including a role for the digital twins developed under the HYPERGRYD project. 2.7 Relation to other activities This report is based on the work performed under Task 3.4. which feeds on tasks 1.4 and 3.2, as well as related ICT architecture and use case development tasks in WP4 - HYPERGRYD Digital Twin Platform as a service, and tasks performed in WP5 -TRL5 demonstration in living labs and virtual labs in LEC, especially Task 5.5 Model demonstration and validation of the HYPERGRYD ICT services at SONNE. The final version of the report (“D3.6: Description and report of energy exchange and smart trading tool for LEC including simulation studies for living labs as active LECs. Final version”) will include the feedback from SONNE users and other stakeholders, provided as part of final activities of T5.5, as well as in related dissemination and exploitation activities in WP6 - Communication and Dissemination. D3.5 Description and report of energy exchange and smart trading tool including simulation studies for living labs as active LECs. Preliminary version 17 2.8 Structure ●Section 1: Executive Summary ●Section 2: Introduction (Scope, Audience, Definitions/Glossary, Abbreviations, Partner Contribution, Baseline, Relation to other activities, Structure) ●Section 3: Heat Pump, Virtual Heat Pump and Heat Demand Digital Twin Development and Integration in the Grid Singularity Exchange ●Section 4: Heat Pump Development Testing via Grid Singularity Exchange Simulation ●Section 5: Grid Singularity Simulation Tool Demonstration: Study of Thermal Asset Coupling and Flexibility in a Sonnenplatz Local Electricity Market D3.5 Description and report of energy exchange and smart trading tool including simulation studies for living labs as active LECs. Preliminary version 18 3 Heat Pump, Virtual Heat Pump and Heat Demand Digital Twin Development and Integration in the Grid Singularity Exchange As part of the HYPERGRYD Project [1], funded by the European Union’s H2020 Programme under Grant Agreement No. 101036656, GSY has enhanced its Grid Singularity Exchange software [2], including the GSY Simulation Tool [3] by developing and integrating digital twins of the heat pump and district heating (virtual heat pump and heat demand digital twin). 3.1. Heat Pump Digital Twin Development In the Grid Singularity Exchange, the heat pump is modelled as a load, which consumes electricity and generates heat. If the heat pump has a storage option (water tank) then this is accounted for by a dedicated heat pump asset trading strategy [8], which facilitates flexibility trading by leveraging the heat storage capabilities of the water tank. The heat hump places bids for electrical energy ranging from an initial to final buying rate, with prices increasing incrementally within the market slot upon the update interval. Asset owners (or managers) can either set the final rate as a default preferred buying rate or select a preferred buying rate based on a smart trading algorithm. One description of a possible smart trading strategy is described below: When the heat pump bids are below the preferred buying rate, the heat pump tries to consume as much energy as possible to satisfy the demand, while also charging the thermal storage for future use, thus maximising the benefit from lower electricity prices (case a). On the other hand, if the electricity price is higher than the preferred buying rate, the heat pump only consumes the required energy to maintain the storage at the same temperature level as the one before the energy trade occurs (i.e. consumes only the energy required to satisfy the asset owner’s heat demand), aiming to minimise the costs incurred by the increased energy prices (case b). a. Bid rate <= Preferred buying rate If the bid price is below the preferred buying rate, the heat pump strategy is to consume the maximum amount of energy. The maximum energy that the heat pump can buy at any point in time is calculated based on the following equation: 𝐸𝑡𝑜𝑏𝑢𝑦= 𝑚𝑖𝑛(𝑃𝑚𝑎𝑥 * 𝑡𝑠𝑙𝑜𝑡, (𝑇𝑚𝑎𝑥− 𝑇𝑐𝑢𝑟𝑟+𝑇𝑑𝑒𝑐𝑟𝑒𝑎𝑠𝑒)* (0. 00116* 𝑉𝑡𝑎𝑛𝑘*ρ𝑤𝑎𝑡𝑒𝑟)/𝐶𝑂𝑃) where - is the maximum power rating 𝑃𝑚𝑎𝑥 - is the slot length 𝑡𝑠𝑙𝑜𝑡 - is the maximum temperature of the storage tank 𝑇𝑚𝑎𝑥 - is the current temperature of the storage tank 𝑇𝑐𝑢𝑟𝑟 - is the temperature decrease of the storage tank due to heat consumption 𝑇𝑑𝑒𝑐𝑟𝑒𝑎𝑠𝑒 - 0.00116 is the energy required to heat one litre of water by 1 °C, in kWh - is the volume of the storage tank 𝑉𝑡𝑎𝑛𝑘 - is the density of water ρ𝑤𝑎𝑡𝑒𝑟 - COP is the coefficient of performance of the heat pump; depends on the heat pump type, and ∆𝑇 = 𝑇𝑐𝑢𝑟𝑟 − 𝑇𝑎𝑚𝑏𝑖𝑒𝑛𝑡 D3.5 Description and report of energy exchange and smart trading tool including simulation studies for living labs as active LECs. Preliminary version 19 b. Bid rate > Preferred buying rate In this situation, the heat pump strategy is to consume the least possible amount of energy in order to keep the temperature at the same level, since the prices are relatively high. Two different situations can arise: - > . In this case, the heat pump does not consume any energy (it’s 𝑇𝑐𝑢𝑟𝑟 𝑇𝑚𝑖𝑛+𝑇𝑑𝑒𝑐𝑟𝑒𝑎𝑠𝑒 not commanded to function because the temperature will be within the temperature limits even if some heat is consumed in the market slot). Therefore, Energy_to_buy = 0 - <= . In this case the heat pump is forced to function, and it 𝑇𝑐𝑢𝑟𝑟 𝑇𝑚𝑖𝑛+𝑇𝑑𝑒𝑐𝑟𝑒𝑎𝑠𝑒 consumes the energy to maintain the temperature at the minimum, taking into account the energy consumption of the user’s premise. In all cases, the maximum power rating is respected, meaning that no more energy is requested than the energy equivalent of the Maximum Power Rating, as illustrated in graphs below (figures 1 and 2). Figure 1. Grid Singularity Exchange Heat Pump Digital Twin: Initial, Final and Preferred Buying Rate in relation to Minimum and Maximum Energy Consumed Figure 2. Grid Singularity Exchange heat pump digital twin example of function to maintain temperature within comfort limits, with the horizontal axis showing timestamps for the selected date and the vertical axis showing temperature in °C. The parameters and default values used to create the template heat pump digital twin are presented in Table 1 below. D3.5 Description and report of energy exchange and smart trading tool including simulation studies for living labs as active LECs. Preliminary version 20 Table 1. Heat Pump Configuration Parameters in the Grid Singularity Exchange Heat Pump Parameters Description Uni t Data Type Manda tory User Input Default Value Name Name of the asset - Text Yes N/A Heat Pump Type Heat pump type specification from drop-down menu: - Air-to-Water - Water-to-Water - Category value Yes Air-toWater Minimum Temperature Minimum water temperature of the heat pump storage (provided by the user based on personal preference; if unavailable, set at 25°C since this is the comfort temperature setting for underfloor heating) °C Constant value No 25 Maximum Temperature Maximum water temperature of the heat pump storage (provided by the user based on personal preference; if unavailable, set at 60°C based on analysis of specification sheets for several heat pumps where value ranges between 55 and 65°C) °C Constant value No 60 Tank Volume Volume/capacity of the thermal storage tank based on technical sheet (nameplate) of the storage tank (if unavailable average volume of 500l set as default based on analysis of specification sheets for consumer heat pump storages) l Constant value No 500 Consumptio n Profile Electricity consumption profile of the heat pump (amount of energy the heat pump consumes in kWh to produce heat), to be added by the user kW h Profile /csv file Yes N/A Initial temperature Initial water temperature of the heat pump storage at the beginning of the simulation (provided by the user based on personal preference; if unavailable, set at the same value as the minimum temperature) ºC Constant value No 25 External Temperature Profile Ambient temperature profile at the community location based on historical weather data (user input, which can be derived from Copernicus Climate Change Service or similar service and resampled to the 15-minute or another market slot [9] resolution) °C Profile /csv file Yes N/A Maximum Power Rating Maximum electricity demand by the heat pump based on technical sheet (nameplate) of the heat pump Note: If unavailable, it can be estimated by using the historical data to calculate the maximum electricity consumption over the coldest month for each heat pump in the community separately, since the kW Constant value No 3 D3.5 Description and report of energy exchange and smart trading tool including simulation studies for living labs as active LECs. Preliminary version 21 heat demand during this month is at the highest level, and then deriving the mean value and setting this value as a standard heat pump capacity. Initial buying rate Minimum buying price for the heat pump in each time slot (based on user input, can be set to feed-in-tariff or another value; the default is 0 since most heat pump owners prefer maximum storage / self-consumption) cts/ kW h Constant value No 0 Final buying rate Maximum buying price for the heat pump in each time slot. The user can choose between “User Input” and “Market maker rate”. cts/ kW h Constant value No Market -maker rate Preferred buying rate Set to buy energy at a price that is equal or less than a certain value in each time slot based on a smart trading algorithm or input value that can be (i) Initial buying rate, in case your strategy is just to keep the heat pump operational to satisfy your own demand, or (ii) Feed-in-tariff in case your strategy is intending to maintain a minimum revenue level for another asset like PV, or (iii) Final buying rate, in case your strategy is intending to maximise storage regardless of the cost, or (iv) Average or another value between the feed-in-tariff and the market maker rate to balance cost and supply security preferences. cts/ kW h Constant value (single value or varied based on algorith m) No 15 As a remark, the thermal losses of the water tank are not accounted for separately in the current model but considered as part of the heat demand. To configure a heat pump asset (HP) in the Grid Singularity Exchange backend code, the following lines are to be added to the children's list of one of the areas in the setup file: Asset(name="Heat Pump", strategy=HeatPumpStrategy()) The HeatPumpStrategy parameters can be set as follows: ●maximum_power_rating_kW: default=3; the maximal power that the HP will consume ●min_temp_C: (default=50); minimum temperature of the HP storage. If the temperature drops below this point, the HP buys energy at any cost ●max_temp_C: (default=60); maximum temperature of the HP storage. If the temperature rises above this point, the HP does not buy any energy. ●initial_temp_C: (default=50); initial temperature of the HP storage ●external_temp_C_profile: (mandatory user input); external temperature that influences the efficiency of the HP. If this parameter is selected, the external temperature is constant for the whole simulation run ●tank_volume_l: (default=500); volume of the storage tank ●consumption_kWh: (mandatory user input); constant power value in kWh or provided power profile as a dictionary that follows the supported format, representing the power that the HP consumes when operating. D3.5 Description and report of energy exchange and smart trading tool including simulation studies for living labs as active LECs. Preliminary version 22 ●preferred_buying_rate: (default=15); rate in cts/kWh that determines the trading strategy; ●source_type: (default: HeatPumpSourceType.AIR); set how the heat exchange is conducted, either via air or water/ground, as it determines the COP calculation; ●order_updater_parameters: of type HeatPumpOrderUpdaterParameters. A template configuration can be seen below. The heat pump price configuration is set as follows: In order to configure the heat pump bid pricing, the order_updater_parameters should be set by assigning the HeatPumpOrderUpdaterParameters data class: from gsy_e.models.strategy.heat_pump import HeatPumpOrderUpdaterParameters from pendulum import duration from gsy_framework.enums import AvailableMarketTypes Asset(name="Heat Pump", strategy=HeatPumpStrategy( order_updater_parameters={ AvailableMarketTypes.SPOT: HeatPumpOrderUpdaterParameters( update_interval=duration(minutes=1), initial_rate=20, final_rate=30)})) The order_updater_parameters expects a dictionary with AvailableMarketTypes as key and HeatPumpOrderUpdaterParameters as values. If not set by the user, the following default values are used: ●update_interval: interval that represents the time between updates of the bid prices performed by the applied trading strategy (default: 1 minute); ●initial_rate: bid rate at the start of the market slot; default value is the feed-in-tariff rate (if not set, the default value is 0 cts/kWh as explained in Table 1 above); ●final_rate: bid rate at the end of the market slot; defined by the infinite bus strategy that is part of the simulation, which is the market maker rate (if not set, the default value is 30 cts/kWh). In addition to backend development supported by the HYPERGRYD project and made available as backend code under the open-source GPL v.3 licence, GSY additionally integrated the development in its user interface (UI), also termed the Singularity Map or the GSY Simulation Tool [3]. In the express heat pump configuration mode, the user is provided with a template (synthetic) heat pump model to include as an asset at a select location in the simulated local energy market by naming the asset and setting its location, while in the advanced heat pump configuration mode, there are two settings referring to its Capacity and Profile and the Trading Strategy, shown in the Figure 3 below. More information on the Grid Singularity heat pump digital twin UI configuration is available in the GSY Wiki [8]. D3.5 Description and report of energy exchange and smart trading tool including simulation studies for living labs as active LECs. Preliminary version 23 Figure 3. Heat Pump Advanced Configuration Options in the Grid Singularity Exchange UI 3.2. District Heating (Virtual Heat Pump and Heat Demand) Digital Twin Development For homes (or other energy community participants) that are connected to district heating (DH), GSY has also developed a digital twin of a heat pump that mimics and replaces the heat demand currently satisfied by a district heating network, modelled as a virtual heat pump (VHP) with storage and a related trading strategy. This VHP implementation [10] is available exclusively in the backend source code, since it will likely be used by energy researchers to simulate how district heating could be replaced by a heat pump to compare their respective performance. Inversely, the VHP can also be used to calculate the heat demand of homes or other energy community participants, and consequently enable the simulation of a district heating connection that would satisfy this heat demand as opposed to electricity assets. Finally, for simulations that simply want to account for the district heating connection without considering a potential replacement with heat pumps, the digital twin of the district heating supply for the measured heat demand can be modelled as a "heat market maker", i.e. digital trading agent with a specific trading strategy [11] representing the district heating provider, which will only sell heat energy to the heat demand digital twin of the respective community member. The selling price of this market maker will be the district heating price that the heat consumer currently pays, in cts/kWh (Euro Cents per kilowatt hour). The heat demand digital twin, in turn, is modelled as a load with a “consumption profile” defined by the measured heat demand in kWh. Thus, the “heat market maker” will only be used in order to cover the heat demand of the heat load and accounting for the monetary cost of heating. 3.2.1 Virtual Heat Pump and Heat Demand Digital Twin Modelling A Virtual Heat Pump (VHP) is akin to a digital twin representing the district heating system role in the electricity market, simulating how the demand for electricity that is satisfied by district heating could be satisfied if a heat pump were used instead. This allows for a comparison of the costs and benefits of the two alternatives, including an assessment on the impact on the grid stability by potentially adding heat pumps to the system. Finally, it facilitates the sizing optimization (the selection of the D3.5 Description and report of energy exchange and smart trading tool including simulation studies for living labs as active LECs. Preliminary version 24 model, rating or other nameplate parameters) of the heat pump if there is a consideration to replace the district heating connection. The modelled trading strategy assumes that the district heating network connection would be replaced by a ground source heat pump, which includes a water tank storage (reference model illustrated in the Figure 4 below). The heat pump is modelled in its entirety as a single system and not as separate units; therefore, the condenser and the compressor are both accounted for in the electricity demand. Other VHP configuration parameters can be modified to suit a particular use case, as described in section 3.2.3 below. Figure 4 : Ground source virtual heat pump diagram The heat pump operation steps are the following: 1. Heat from the water pipes (boreholes) in the ground (#1 in Figure 4) is transferred to the source side of the heat pump, via the heat exchanger (evaporator) (#2 in Figure 4) 2. The heat pump compressor (#3 in Figure 4) increases the pressure and temperature of the gas medium. 3. The heat from the condenser side of the heat pump is transferred to the water tank storage via the second heat exchanger (#4 in Figure 4) and the gas medium is cooled and converted to liquid (condensed). 4. The water tank storage (#5 in Figure 4) temperature is increased as a result of the heat produced by the heat pump condenser in step #3. 5. Heat from the water tank storage is used in order to provide the heat demand of the building (#6 in Figure 4). Unlike the heat pump model described above, district heating networks use water pipes that are directly connected to the building, and usually the supply temperature, return temperature and water flow rates are measured in order to facilitate customer billing. The virtual heat pump model can use these district heating measurements to calculate the heat demand of the building. The heat demand is then provided as an input for the VHP model to calculate the energy that needs to be consumed by the heat pump in order to provide the heat demand, taking into account the flexibility that the water tank storage provides. 3.2.2. Virtual Heat Pump Model Equations The model of the virtual heat pump comprises a set of mathematical equations that each simulates a different component of a heat pump system. The system of linear equations is optimised, in order to D3.5 Description and report of energy exchange and smart trading tool including simulation studies for living labs as active LECs. Preliminary version 25 4.2. Heat Pump Digital Twin Test Results 4.2.1. Heat Pump Digital Twin Measurement Test Results Each asset participating in the simulated SONNE LEM was sized and configured according to the nameplate specifications of the real assets. Plots and graphs of the asset data inputs are shown in figures generated by the web-based GSY Simulation Tool [3], validating that their values are within a specific acceptable range. a) Home 5 Measurement Results Figure 6. SONNE Home 5 asset overview in the Grid Singularity simulation interface - 1 Load “05.0.05”: - initial buying rate: 0 cts/kWh - final buying rate: 30 cts/kWh - price update interval: 1 minute - 1 PV “Custom PV”: : - initial selling rate: 30 cts/kWh - final selling rate: 0 cts/kWh - price update interval: 1 minute - local generation profile - capacity: 40 kWp - orientation: 235° - tilt: 18° - 1 Battery “Battery” (nameplate values taken from [21]) : - initial buying rate: 0 cts/kWh - final buying rate: 25 cts/kWh - initial selling rate: 30 cts/kWh - final selling rate: 25.1cts/kWh - price update interval: 1 minute - capacity: 11.52 kWh - maximum power 11.52kW - initial capacity: 4% - minimum SOC: 4% - 1 Heat pump “05.0.03”: D3.5 Description and report of energy exchange and smart trading tool including simulation studies for living labs as active LECs. Preliminary version 32 - water-to-water - minimal temperature: 30 °C - maximum temperature: 45 °C - initial temperature: 30 °C - tank volume: 1000 l - maximum power rating: 2.78 kW - initial buying rate: 0 cts/kWh - final buying rate: 30 cts/kWh - preferred buying rate: 30 cts/kWh Figure 7. SONNE Home 5 heat pump configuration in the Grid Singularity simulation interface b) Home 9 Measurement Results Figure 8. SONNE Home 9 asset overview in the Grid Singularity simulation interface - 4 loads (“09.0.02”, “09.0.03”, “09.0.04”, “09.0.05”): - initial buying rate: 0 cts/kWh - final buying rate: 30 cts/kWh - price update interval: 1 minute - 1 heat pump “09.0.01”: D3.5 Description and report of energy exchange and smart trading tool including simulation studies for living labs as active LECs. Preliminary version 33 - water-to-water - minimal temperature: 25 °C - maximum temperature: 30 °C - initial temperature: 25 °C - tank volume: 400 l - maximum power rating: 1.26 kW - initial buying rate: 0 cts/kWh - final buying rate: 30 cts/kWh - preferred buying rate: 30 cts/kWh Figure 9. SONNE Home 9 heat pump configuration in the Grid Singularity simulation interface c) Home 10 Measurement Results Figure 10. SONNE Home 10 asset overview in the Grid Singularity simulation interface - 1 load “10.0.02”: - initial buying rate: 0 cts/kWh - final buying rate: 30 cts/kWh - price update interval: 1 minute D3.5 Description and report of energy exchange and smart trading tool including simulation studies for living labs as active LECs. Preliminary version 34 - 1 PV “10.0.04”: - initial selling rate: 30 cts/kWh - final selling rate: 0 cts/kWh - price update interval: 1 minute - 1 heat pump “10.0.01”: - water-to-water - minimal temperature: 15 °C - maximum temperature: 50 °C - initial temperature: 15 °C - tank volume: 600 l - maximum power rating: 1.56 kW - initial buying rate: 0 cts/kWh - final buying rate: 30 cts/kWh - preferred buying rate: 30 cts/kWh Figure 11: SONNE Home 10 heat pump configuration In the Grid Singularity simulation interface After sizing and configuring the assets, a simulation of a local energy market (LEM), mimicking an operation of peer-to-peer trading in an energy community, was undertaken to validate that the generated and consumed energy values correspond to the relevant smart metre and sensor data provided by SONNE. Figures below (12a compared to 12b, and 13a compared to 13b) show that the raw electricity consumption and production profiles provided for the SONNE Home 5 Load and Home 10 PV, respectively, match the trading profiles that were generated based on the provided configuration parameters. These also show that the correction of the raw data for missing and shifted timestamps was performed correctly. D3.5 Description and report of energy exchange and smart trading tool including simulation studies for living labs as active LECs. Preliminary version 35 Figure 12a. SONNE Home 5 energy consumption profile on 1-7 July 2023 plotted from original raw data with the horizontal axis showing timestamps for the selected date and the vertical axis showing consumed energy in kWh Figure 12b. SONNE Home 5 energy trading profile on 1-7 July 2023 extracted from the Grid Singularity simulation tool with the horizontal axis showing timestamps for the selected date and the vertical axis showing energy bought in kWh Figure 13a. SONNE Home 10 PV production profile on 1-7 July 2023 plotted from original raw data with the horizontal axis showing timestamps for the selected date and the vertical axis showing produced energy in kWh D3.5 Description and report of energy exchange and smart trading tool including simulation studies for living labs as active LECs. Preliminary version 36 Figure 13b. SONNE Home 10 PV trading profile on 1-7 July 2023 plotted from original raw data extracted from the Grid Singularity simulation tool with the horizontal axis showing timestamps and the vertical axis showing energy sold in kW Finally, the heat pump coefficient of performance (COP) value was validated. The average water tank temperature per tank volume was calculated based on the energy traded by the heat pump at a specific time and the resulting tank temperature. This was then checked against the heat pump nameplate parameters, evaluating whether the resulting temperature in the simulation remained within the nameplate minimum / maximum range. Figure 16 below shows the correlation of the COP with the ambient temperature (wave form) and the storage temperature. The peaks in the graph are a result of a decreased storage temperature and the subsequent unmatched electricity demand of the heat pump. Once affordable energy is available (or the temperature drops to the minimum storage temperature), the available energy is used to compensate for the storage temperature drop during the past time period, leading to temporarily higher COP values. Another observation is that the COP values for Home 9 (shown in red) are higher than for Home 5 (shown in blue) and Home 10 (shown in green) due to the different nameplate characteristics in the modelled heat pumps. Home 9 has a heat pump with significantly lower maximum power rating than the other homes’ heat pumps, obliging it to operate with a higher COP factor to satisfy the household’s heat demand. Figure 14. Coefficient of performance as a function of time for heat pumps in simulated SONNE LEM for the period 1-7 July 2023 with the horizontal axis showing timestamps for the selected week and the vertical axis showing COP factor 4.2.2. Heat Pump Digital Twin Trading Test Results Finally, the GSY simulation of a local energy market (LEM) for a partial SONNE community was used to validate whether the trading behaviour of the assets corresponds to the size / capacity of the assets, as well as to validate the assets’ trading strategy. D3.5 Description and report of energy exchange and smart trading tool including simulation studies for living labs as active LECs. Preliminary version 37 Figure 15 below shows to what extent the 3 community members used the self-produced energy and how much energy had to be bought from an external supplier i.e. utility (represented by the “Grid Market” entry in the bills table shown in Figure 15). The results are aligned with the individual community member results presented below for each participating home and for all heat pump assets. In total, 1482.07 kWh of electricity was produced by the community, and an additional 156.45 kWh provided externally by the utility (shown in the Grid Market “Sold” column). Most of the communityproduced energy was bought by the external market (1087.25 kWh, shown in the Grid Market “Bought” column). The significant export results in a low self-consumption of 28.2%. The self-sufficiency is quite high at 73.2% because most of the demanded energy of 584.23 kWh was provided by the community members. Figure 15. Results in the Grid Singularity simulation interface showing community energy traded and bills table and the self-sufficiency and self-consumption rates Prosumer homes 5 and 10 generate revenue based on the PV generation. In contrast, Home 9 is an energy consumer only with a consumption load and a heat pump. The results are presented and explained for each home below. D3.5 Description and report of energy exchange and smart trading tool including simulation studies for living labs as active LECs. Preliminary version 38 4.2.2.1 Home 5 Trading Results Figure 16. SONNE Home 5 energy bills and the self-sufficiency and self-consumption rates In the Grid Singularity simulation interface The bills in Home 5 in Figure 16 show a self-sufficiency rate of 67.2%, indicating that the majority of the energy demand was satisfied by own PV generation. The Home 5 PV produced 1304.4 kWh which was mostly sold outside of the home market which resulted in a low self-consumption rate of 12.9%. This is a result of the load profile also demanding energy when there is no PV production (mainly at night). Furthermore, the storage did not sell all its energy to the Home 5 load, because the energy offer is also made to other community participants, which results in trades with loads in other homes. If the energy offer was restricted to prioritise loads in Home 5, the self-consumption and self-sufficiency rates would be higher. 4.2.2.2. Home 5 Heat Pump Trading Results The heat pump in Home 5 was set up to always consume as much energy as required in order to store as much heat as possible in the storage tank. This was achieved by setting the preferred buying rate to be equal to the final buying rate. In the energy bills and traded energy table, the total energy demand of the heat pump in Home 5 (05.0.03) is 1.2kWh, which is comparatively low compared to other heat pumps (e.g. in the other houses. Figure 17 shows the trading profile of the heat pump in Home 5. D3.5 Description and report of energy exchange and smart trading tool including simulation studies for living labs as active LECs. Preliminary version 39 Figure 17. SONNE Home 5 heat pump energy trading profile in the Grid Singularity simulation interface with the horizontal axis showing timestamps for the selected week and the vertical axis showing energy bought in kWh Green bars show the demanded consumption of the heat pump and red bars show the actual bought energy. The graph shows that there are periods of time (time slots) when the heat pump demanded energy but did not buy it. This is a result of both the price setting of the heat pump and the behaviour of the energy market. The heat pump was configured to submit bids for energy starting at a rate of 0 cts/kWh at the beginning of the market slot and increasing to the market maker rate (set to 30 cts/kWh) by the end of the market slot. Consequently, there may not have been sufficient time to clear these bids at the last tick in the market slot (in the Grid Singularity Exchange tools, the lowest time unit is called “tick” and its length can be configured by the user; in this simulation the default tick length of 15 seconds was used). Generally, if a bid is matched with an offer in a lower market, all connected offers are split along the trading path. The corresponding offers in the other branches of the setup tree will be deleted instantly, but the reposting of the market maker offer in that branch's markets will take some time (2 ticks or 30 seconds). For example, if the heat pump in Home 10 purchases some of the offered energy from the utility provider in the last tick of the market slot in the Home 10 market, there will be no offer from the utility provider in Home 5 because it was deleted immediately after the trade in Home 10. If there are no other sources of energy, no energy offers will be left in the Home 5 market. This will leave the Home 5 heat pump bid unmatched, and no electricity will be converted into heat. Consequently, the temperature of the heat pump storage drops. This can be observed in the storage tank temperature graph during the observed times, shown in Figure 18. D3.5 Description and report of energy exchange and smart trading tool including simulation studies for living labs as active LECs. Preliminary version 40 Figure 18: SONNE Home 5 heat pump storage tank temperature time series in the Grid Singularity simulation interface with the horizontal axis showing timestamps for the selected week and the vertical axis showing the tank temperature in °C When the heat pump is once again able to purchase energy, the temperature begins to rise. These temperature drops correspond to nighttime periods when neither PV-generated nor stored energy is available. During this time, the storage tank temperature drops because there is no energy available between 12 PM and 5 AM. The temperature drop is only very small (maximal 0.1°C to a minimum 44.9°C) because the low energy demand faces a large storage capacity of 1000 litres. Due to the random dispatching of offers in the Grid Singularity simulation exchange, even in these times, the Home 5 heat pump is able to buy energy which results in an increase of the storage tank temperature. To determine if the pricing settings are causing this behaviour, we conducted another simulation run. In this run, the final buying price of the heat pump was set higher than the market maker rate (at 35 cts/kWh). This adjustment allowed the heat pump to reach the market maker rate earlier in the market slot. With this change, the heat pump was always able to purchase energy resulting in no unmatched demand and no drops in storage tank temperature. 4.2.2.3 Home 9 Trading Results In Home 9, only loads and one heat pump are connected. Consequently, the self-sufficiency and the self consumption rates are 0, as shown in the figure below: D3.5 Description and report of energy exchange and smart trading tool including simulation studies for living labs as active LECs. Preliminary version 41 Figure 23. Sonnenplatz Großschönau GmbH digital twin configuration showing number and type of LEC participants and energy assets Project partner, KTH Royal Institute of Technology, was tasked with retrieving real-time data of all of the SONNE energy metres and operation sensors via the Internet by means of Message Queuing Telemetry Transport (MQTT) protocol since the previously collected historical data from SONNE was not of sufficient quality for simulation purposes. The resulting time-series data, starting with 1 April 2023, was uploaded in the form of an Excel file on the project’s electronic database (Sharepoint) for use by Grid Singularity (GSY), together with additional data collected directly from SONNE by GSY, to test the developed heat pump digital twin and configure the SONNE LEC to conduct simulations to demonstrate the project’s developments. As envisaged, GSY did not use any personal data pertaining to community members other than anonymised, non-specific geolocation of respective energy assets. For the purpose of this study, the SONNE management has the role of a data controller and KTH and GSY have acted as data processors, respecting the GDPR regulation [14]. 5.1.2. Demonstration Study Energy Asset Data Detailed information on the SONNE historical data used to configure the LEM (energy community) participants and their energy assets for this demonstration study is presented in a table in Annex I, which is not published with this report to ensure GDPR compliance. The report instead includes a high level overview of the SONNE asset data here below. The data requirements for the GSY Simulation Tool [3] are available on the relevant page of the Grid Singularity Wiki [13]. a) Electricity consumption: 23 SONNE load assets, all with a metered consumption load profile, were configured, with an example profile shown in Fig. 2 below: D3.5 Description and report of energy exchange and smart trading tool including simulation studies for living labs as active LECs. Preliminary version 48 Figure 24. Load consumption profile of SONNE Building 1, in Wh for 1-7 August 2023 with the horizontal axis showing timestamps for the selected week and the vertical axis showing the energy consumed in Wh b) PV generation: A total of nine (9) buildings in the SONNE have PV assets, with four (4) metered and five (5) non-metered. To configure the non-metered PVs, GSY deployed the Grid Singularity’s Custom PV tool [16], which models PV generation by leveraging location-based data profiles generated from the Energy Data Map [17] provided by rebase.energy [18] based on the PV’s capacity, location, azimuth, and tilt. It is important to note that some of the PVs in the community, such as those in buildings 8 and 12, were set to be purely feed-in, due to an old tariff model that made it economically more viable to feed-in the entire generated electricity. The total PV generation profile in the SONNE for the week of 1-7 August 2023 is shown in Fig. 3 below: Figure 25. Total PV generation in the SONNE in kW for 1-7 August 2023, with the horizontal axis showing timestamps for the selected week and the vertical axis showing the power in kW c) Heat demand - A notable feature of the SONNE community is its district heating network, which satisfies the community’s heat demand, measured in kWh. To assess the impact of heat pumps and subsequently the flexibility that could potentially be provided by thermal storage, virtual heat pumps with storage were sized and simulated using the Grid Singularity's Virtual Heat Pump Model [10] described in Section 3 of this report. Although heat demand D3.5 Description and report of energy exchange and smart trading tool including simulation studies for living labs as active LECs. Preliminary version 49 measurements were listed as available data points, these were not provided and the heat demand of each building was instead calculated by leveraging the district heating measurements, namely source/return temperature and water flow, which were provided as 15-minute time series profiles. For one of the required values, the ambient temperature profile at the community location, the historical weather data was derived from the Copernicus Climate Change Service [15] and resampled to the 15-minute market slot resolution. In the SONNE community, six (6) homes/buildings with 14 out of 15 available district heating network connections were simulated as virtual heat pumps (the heat demand of the Commercial Building 6 was not included in the simulation because the data pertaining to the district heating network was not available and could not be extrapolated; importantly, this did not impact the results since the same approach was applied to the base case scenario). The capital cost (CAPEX) of purchasing and installing the heat pumps was not considered in the financial evaluation of the heat pump impact compared to using the district heating exclusively since a reliable CAPEX cost analysis would require a simulation of a dataset with a longer timeframe to produce reliable results. d) Energy storage / batteries - One of the significant advantages of battery storage is its capacity to increase the self-consumption rates of rooftop photovoltaic (PV) systems, thereby curbing reliance on conventional grid power. In the SONNE community, the 11 available battery storages were configured. e) Heat pumps - A heat pump is a device used to heat water and premises by extracting heat from one place (air source, water source, and ground source/geothermal) and moving it to another (some pumps also have a cooling function but only heating is considered here). Depending on the external temperature and the heating needs, a heat pump usually has a coefficient of performance (COP) of around 2-4 [6], which means that it is able to produce 2-4 kWh of heat by consuming 1 kWh of electricity. Since the SONNE has three (3) water-to-water heat pumps with storage but the installed heat pump models do not provide the measurements related to heat demand and only provide the heat pump’s total electricity consumption, it is not possible to simulate the storage function of these heat pumps. If the respective homes/buildings were connected to the district heating network, we could have used the heat demand measurements or water flow, supply and return temperature data to synthetically generate the heat demand but this is not the case. Therefore, these heat pumps were modelled only as consumers of electricity (loads), with the Fig. 4 below showing the daily consumption profile for the three simulated SONNE heat pumps: D3.5 Description and report of energy exchange and smart trading tool including simulation studies for living labs as active LECs. Preliminary version 50 Figure 26. Combined daily consumption profile in kWh of 3 SONNE heat pumps on 1 August 2023 with the horizontal axis showing timestamps per 15-minute slot for the selected day and the vertical axis showing the energy consumption in kWh In addition to contributing effort to synthesise the required data where actual data was only partially available, significant effort was dedicated to data quality review and processing, including input formatting for consumption and production profiles, in order to correct the misalignment of the timestamps with the slot start times. For example, depending on the profile received, there was a time shift of 1 to 3 minutes that needed to be adjusted and additionally reformatted according to the GSY Simulation Tool data format requirements [13]. Moreover, there were several gaps (missing data) in provided consumption profiles in the range of minutes, which were resolved by completing the gaps with the latest consumption value; for example if the consumption value for 10:00 AM was missing, the value from 9:45 AM was inserted. The reason for these discrepancies lies in insufficient measurement quality which is still a frequent occurrence when it comes to energy data. Furthermore, incorrect data was identified in several time slots for district heating measurements in Commercial Building 4 and Office 8, where the district heating supply temperature value was higher than the return temperature value, which is not physically possible. SONNE management assumption is that this outcome can be explained either as an after-reaction of the metre or as a consequence of accumulated heat, at a time when the water flow starts after standing still for a while. To correct for errors, the supply temperature was set to be equal to the return temperature in these instances. Other data processing undertaken included: - The water flow data received was instantaneous m^3 per hour which was converted to cumulative m^3 per 15 minutes; - To calculate the maximum power ratings of the battery, the nominal voltage was multiplied with the maximum output current as this data was not available in the provided battery manuals; - Consumption data received from the heat pump in building 5 were impulse time series data that reported values only when the accumulated energy consumption of each metre surpassed 0.1 kWh (approximately once every 12 hours). These had to be converted into non-accumulated time series by linearly interpolating each measurement, thereby assuming that the consumed energy was constant for the time slots with no impulse data D3.5 Description and report of energy exchange and smart trading tool including simulation studies for living labs as active LECs. Preliminary version 51 - In Hotel 7 and Residential Buildings 10 and 11, the loads and the PV were behind a single metre. In other words, the PV was metered, but aggregated at the home/building level. The metered data therefore needed to be disaggregated using the Grid Singularity PV tool to calculate the actual PV generation, and then combined with the metered data to extrapolate the values for self-consumption, load profile and export from the grid. Specifically, the self-consumption was calculated by subtracting the metered PV feed-in from the total yield of the PV panel. The derived self-consumption value was then added to the metered load consumption from the grid profiles, based on the formulae below: 𝑆𝑒𝑙𝑓 𝐶𝑜𝑛𝑠𝑢𝑚𝑝𝑡𝑖𝑜𝑛 = 𝑃𝑉𝑦𝑖𝑒𝑙𝑑− 𝑃𝑉𝑓𝑒𝑒𝑑−𝑖𝑛 𝑇𝑜𝑡𝑎𝑙 𝐿𝑜𝑎𝑑 = 𝐿𝑜𝑎𝑑𝑖𝑚𝑝𝑜𝑟𝑡 𝑓𝑟𝑜𝑚 𝑔𝑟𝑖𝑑 + 𝑆𝑒𝑙𝑓 𝐶𝑜𝑛𝑠𝑢𝑚𝑝𝑡𝑖𝑜𝑛 When comparing scenaria, the following formula was applied, adapted for the scenario number and the analysed parameters (net cost, self-sufficiency rate, etc): 𝑛𝑒𝑡 𝑐𝑜𝑠𝑡_𝑜𝑓_𝑠𝑐𝑒𝑛𝑎𝑟𝑖𝑜 𝑣.2 − 𝑛𝑒𝑡 𝑐𝑜𝑠𝑡_𝑜𝑓_𝑏𝑎𝑠𝑒_𝑐𝑎𝑠𝑒_𝑠𝑐𝑒𝑛𝑎𝑟𝑖𝑜 / 𝑛𝑒𝑡 𝑐𝑜𝑠𝑡_𝑜𝑓_𝑏𝑎𝑠𝑒_𝑐𝑎𝑠𝑒_𝑠𝑐𝑒𝑛𝑎𝑟𝑖𝑜 A final note with regards to SONNE asset data is that there was an ex post identification of metering errors due to issues with metre impulse readings and incorrect PV capacity information supplied by two PV owners. Nevertheless, since all results involve a comparison with the base case, the conclusions drawn from this study remain correct. 5.1.3. Demonstration Study Market Mechanism and Pricing Data f) Electricity prices and grid tariffs: In the SONNE community, the electricity prices for households in 2023 ranged between 19 and 32 cents per kilowatt-hour (kWh). The grid fees amounted to approximately 19% of that value, energy price 59.4% and taxes and surcharges 21.7%. To simplify the simulation set-up, a selling rate of 25.5 cts/kWh was used as this was the mean value for the utility rate, also termed the Market Maker rate. The specific feed-in-tariff for the residential photovoltaic (PV) systems was between 13-27 cent/kWh in the simulated period, and a mean of 20 cts/kWh was used in the simulation. g) Two time-of-use (ToU) grid tariffs were implemented for simulation scenarios v.5 and v.6, respectively. The first is a day ToU tariff, where electricity prices are lower during the day when PV production is high, leading to increased sales of PV generation to the grid / external market and reducing self-consumption. In contrast, the night ToU tariff has high electricity prices during the day when PV production is high, incentivising all community participants to buy electricity from community-produced PV to the extent possible, increasing self-consumption. The example profiles for the two different ToU tariffs are shown in Figures 27a and 27b below. D3.5 Description and report of energy exchange and smart trading tool including simulation studies for living labs as active LECs. Preliminary version 52 Figure 27a. Example of Day Time-of-Use Tariff applied in the SONNE LEC simulation with the horizontal axis showing timestamps for the selected day and the vertical axis showing the price €cents/kWh Figure 27b. Example of Night Time-of-Use Tariff applied in the SONNE LEC simulation with the horizontal axis showing timestamps for the selected day and the vertical axis showing the price €cents/kWh h) Heat prices charged by the district heating company: According to the information provided by the SONNE management, the district heating price for the simulated period is 12.24 cents/kWh including taxes. This price was used in the calculation of the heating energy bills in the simulation scenario v.1, which simulated the base case where the heat demand was fully satisfied by the district heating. i) Market clearing mechanism - To simulate peer-to-peer trading in several simulation scenarios, a two-sided pay-as-bid [20] market clearing mechanism was applied by the GSY simulation tool, which is the trading mechanism most commonly used by other peer-to-peer energy communities, thus rendering the results more widely applicable. . The market pricing for external supply (utility rate) and surplus sale to the grid (feed-in-tariff) and network cost (grid fees) reflect the actual conditions in SONNE. To incentivise local trading, a delay of 30 seconds was implemented when forwarding bid/offer to the grid than when propagating the bids/offers to the community. As a D3.5 Description and report of energy exchange and smart trading tool including simulation studies for living labs as active LECs. Preliminary version 53 result, the assets in the community have a greater chance of trading energy amongst themselves before any deficit and surplus is bought or sold from or to the grid. To conclude, the SONNE was digitally configured as a local energy market (LEM or an energy community) in the GSY Simulation Tool [3], with the geolocation in Großschönau, Austria (latitude: 48.64, longitude: 14.93) and digital twins of select 13 LEM participants with 64 different energy assets. The energy assets’ configuration [19] was implemented using the Grid Singularity Exchange backend code [2] based on the historical electricity consumption and generation data, with supplemented synthetic data where needed, as described above, with the set-up file excerpt shown in the Figure 28 below. Figure 28. Segment of the Grid Singularity Exchange configuration setup file for the simulation scenario v.1 5.1.4 Demonstration Study Simulation Scenarios The study served to validate the functionality of the developed solution, assessing the role of heat pumps and district heating in local energy markets, the impact of flexibility assets and otherwise evaluating the benefits of sector coupling and peer-to-peer energy trading for individual consumers and prosumers organised in energy communities. Seven simulation scenarios with different configurations of a digital twin of a local energy market (an energy community) based on SONNE data have been defined and presented in the table below, including the implementation of two tariff models in scenario 7. To accurately assess the impact of these heat generation units in various D3.5 Description and report of energy exchange and smart trading tool including simulation studies for living labs as active LECs. Preliminary version 54 weather conditions, two distinct simulations were conducted for each of the defined scenarios, corresponding to summer and winter temperature conditions, by selecting the first week of August 2023 and the first week of November 2023 from the SONNE dataset, respectively. To conclude, a total of 18 different simulations were conducted for the SONNE study using the GSY Simulation Tool. The demonstration study objectives, similar to validation testing simulations, have been evaluated using the following GSY Simulation Tool key performance indicators [5], focusing on economic and environmental criteria and indirectly also addressing the social criteria, as envisaged by the outcomes of T1.4, described in the project deliverable D4.1 HYPERGRYD Use Case definition including stakeholders and Roles for DT PaaS [1]: ●Self-consumption rate ●Self-sufficiency rate ●The energy cost savings ●The net energy traded. Two specific expected results (ER) were evaluated in the demonstration study, as envisaged by the EU Horizon 2020 HYPERGRYD project [1], specifically by the “D7.2 Final Innovation Report and Exploitable Results characterization.” ●ER1: Integration of flexible heating systems in electrical grid ●ER2: Dynamic pricing effect on flexible assets At the same time, the study also provided evidence for the following expected results, since it included a developed digital twin of the heat pump and the virtual heat pump model: ●ER9: Grid Singularity simulation tool enhancement ●ER10: Software enhancements of open-source Grid Singularity Exchange. Table 2. SONNE LEC simulation scenarios and objectives, GSY simulation study for HYPERGRYD, 2023-2024 SONNE LEC Simulation Scenarios (Digital Twin Configurations of SONNE as Local Energy Community) SONNE Simulation Study Objectives Scenario v. 1: Base case, no P2P trading Status quo, with all available assets and the district heating network connection configured and no peer-to-peer (P2P) electricity trading To define base case scenario with the current LEM asset and market configuration Scenario v. 2: Altered asset configuration (district heating replaced by heat pumps), no P2P trading All available assets configured but the district heating network connection is replaced with additional heat pumps; no P2P electricity trading To evaluate the benefit of heat pumps in a legacy electricity market by assessing the LEM performance and grid stability if district heating were to be replaced by heat pumps D3.5 Description and report of energy exchange and smart trading tool including simulation studies for living labs as active LECs. Preliminary version 55 Scenario v. 3: Base case with activated P2P trading All available assets configured to reflect the base case but with activated P2P electricity trading based on a pay-as-bid market clearing mechanism To evaluate the impact of P2P trading on the LEM without changing the asset configuration Scenario v.4 : P2P trading with heat pumps replacing district heating All available assets configured but the district heating network connection is replaced with additional heat pumps; P2P electricity trading activated To evaluate the impact of P2P trading on the LEM with additional flexibility provided by heat pumps replacing district heating and to additionally assess the role of batteries in enhancing the flexibility. Scenario v. 5: Base case with activated P2P trading and applied dynamic tariffs All available assets configured to reflect the base case but with activated P2P electricity trading and applied time-of-use tariffs To evaluate the impact of P2P trading on the LEM with optimised flexibility management via dynamic tariffs Scenario v. 6: Activated P2P trading with applied dynamic tariffs and heat pumps replacing district heating All available assets configured but the district heating network connection is replaced with additional heat pump and P2P trading is activated with applied time-of-use tariffs To evaluate the impact of P2P trading on the LEM with additional flexibility provided by heat pumps replacing district heating and optimised flexibility management via dynamic tariffs Scenario v.7 : Activated P2P trading with an optimised heat pump trading strategy and heat pumps replacing district heating As in scenario v.4 all available assets configured to reflect the base case but the district heating network connection is replaced with additional heat pumps, with activated P2P electricity trading; additionally, an optimised trading algorithm is implemented To evaluate the impact of P2P trading on the LEM with optimised trading strategies for flexible heat pump assets 5.2 Demonstration Study Results Presentation and Analysis 5.2.1. Base case scenario (SONNE LEC v.1) In this base case scenario, the digital twin of the SONNE as a local energy community (SONNE LEC v.1) was configured in the GSY Simulation Tool [3] to reflect the current conditions in terms of available assets (modelled based on provided historical and supplementary data as described above) and market conditions where there is no peer-to-peer energy trading (community members are only D3.5 Description and report of energy exchange and smart trading tool including simulation studies for living labs as active LECs. Preliminary version 56 self-consuming and selling any surplus only to the grid, with the utility serving as the only electricity external supplier). To simulate Base Case scenario v.1, the district heating network connection is configured in a way that reflects the current pricing of the community. In this configuration, the digital twin of the district heating supply for the measured heat demand is modelled as a "heat market maker," acting as a digital trading agent with a specific trading strategy [11] representing the district heating provider. Its role is to sell heat energy exclusively to satisfy the heat demand of the respective community member. The selling price of this market maker is the district heating price that the heat consumer currently pays, which is 12.24 euro cts/kWh as noted above. The heat demand digital twin, in turn, is modelled as a load with a “consumption profile” defined by the measured heat demand in kWh. Thus, the “heat market maker” is utilised solely to meet the heat demand of each community member and account for the associated heating expenses. The resulting costs for community members are then added to their final energy bills. 5.2.1.1. Self-Consumption and Self-Sufficiency of SONNE LEC v.1 The self-sufficiency and self-consumption rates for the base case v.1 for the first week in August and November 2023, respectively, are shown in percentage terms in Table 3 below. In August, each building primarily self-consumes. In Hotel Building 7, Office Building 8 and Residential Building 3, the self-consumption values are especially high, at 95.9%, 96.2% and 99.9% respectively, with almost all of the energy generated by the PV modules in these buildings consumed to satisfy their own demand. Hotel Building 7 and Residential Building 8 also have battery storages to further bolster their self-consumption. Similar behaviour is exhibited in the November results, where the respective self-consumption values for Hotel Building 7, Office Building 8 and Residential Building 3 are 85.5%, 100% and 99.7%, respectively. The differences in self-consumption values are attributed to the difference in electricity demand of each building during the times of renewable energy production. These values are as expected considering that offices are usually closed during the night, while there is higher evening activity in a hotel than in a residential building. In comparison to August, the aggregate community results for November exhibit significantly lower self-sufficiency (27.6% as opposed to 43.6%, which 36.70% decrease) but higher self-consumption rates (47.3% as opposed to 36.1%, which is 31.02% increase). This is a consequence of the reduction in the PV energy production during November due to lower solar radiation, combined with increased heat demand of the buildings during the winter period, compared to the first week of August where heat demand was minimal due to favourable weather conditions. Table 3. Self-sufficiency and self-consumption rates for SONNE LEC and per participant, simulation v.1 for 1-7 August and 1-7 November 2023, based on data provided by SONNE Building ID SelfSufficiency August (%) SelfSufficiency November (%) SelfConsumption August (%) SelfConsumption November (%) Commercial Building 1 (load + heat demand + solar non-metered PV 5.4 kWp) 63.8 41 22.6 39.4 D3.5 Description and report of energy exchange and smart trading tool including simulation studies for living labs as active LECs. Preliminary version 57 Commercial Building 6 24.9 0 24.9 Hotel Building 7 108.9 5.3 103.5 Office Building 8 214.7 0 214.7 Residential Building 1 60.9 0 60.9 Residential Building 2 75 5.4 69.7 Residential Building 3 17 0 17 Residential Building 4 155 28.4 126.6 Residential Building 5 5.1 0 5.1 SONNE LEC Aggregate 1330.2 179.7 1150.5 5.2.2.3. Energy Import and Export of SONNE LEC v.2 A graph of the energy imports and exports in the SONNE LEC v.2 scenario is shown in Figures 30a and 30b below for the two simulated periods. A somewhat lower level of exported energy is observed in the summer period compared to the base case v.1 (from 3508.1kWh to 3319.0 kWh, which represents a 5.4% decrease), as well as a higher level of imported energy (from 2754.6 kWh to 3521.9 kWh, which represents a 27.9% increase). In the winter period when the solar radiation is low there is a significantly lower level of exported energy (values reduced from 1146.2 kWh to 898.7 kWh, which represents a 21.6% decrease), and a dramatically higher level of imports to satisfy the increased heat demand (from 3202.8 kWh to 5216.6 kWh, which represents a 62.9% increase). Figure 30a. Net energy import and export graph for the SONNE LEC, simulation v2 for 1-7 August 2023 with the horizontal axis showing the building IDs and the vertical axis showing exported and imported energy in kWh D3.5 Description and report of energy exchange and smart trading tool including simulation studies for living labs as active LECs. Preliminary version 64 Figure 30b. Net energy import and export graph for the SONNE LEC, simulation v.2 for 1-7 November 2023 with the horizontal axis showing the building IDs and the vertical axis showing exported and imported energy in kWh 5.2.3. Base case with activated P2P trading (SONNE LEC v.3) In the SONNE LEC v.3 scenario, the current conditions of SONNE are reflected in terms of available assets (modelled based on provided historical data and supplementary data as described above) and the current market conditions in terms of utility and feed-in-tariff rate, but the market mechanism is altered to pay-as-bid peer-to-peer (P2P) trading, enabling LEM participants (community members) to trade surplus (in this case PV-based) electricity generation among each other. 5.2.3.1. Self-Consumption and Self-Sufficiency of SONNE LEC v.3 As shown in Table 7, with the introduction of P2P trading, the community's self-sufficiency levels increased from 43.6% to 97.6% for the simulated week in August (+123.85%), and from 27.6% to 56.5% in November (+104.71%), compared to the base case scenario v.1. Furthermore, the self-consumption values rose from 36.1% to 80.7% (+123.55%) and from 47.3% to 96.9% (+104.86%). in August and November, respectively. Since P2P trading incentivizes local trading, each building in the community self-consumes and either sells surplus or buys the required electricity within the community before interacting with the external grid. Similarly to scenario v.1, when comparing the August and November results, we observe higher self-sufficiency values in August and lower self-consumption values in November, which is attributed to the increased energy demand during the winter months. It is worth noting that the self-sufficiency and self-consumption levels of individual buildings remained unchanged compared to v.1, since the energy production and demand of these buildings have remained constant, and we did not change the installed parameters of the solar panels in buildings 8 and Residential Building 5 that exclusively sell the generated electricity to the external grid. D3.5 Description and report of energy exchange and smart trading tool including simulation studies for living labs as active LECs. Preliminary version 65 Table 7: Self-sufficiency and self-consumption rates for SONNE LEC and per participant, simulation v.3 for 1-7 August/November 2023 Building ID SelfSufficiency August (%) SelfSufficiency November (%) SelfConsumption August (%) SelfConsumption November (%) Commercial Building 1 63.8 41 22.6 39.4 Commercial Building 2 0 0 0 0 Commercial Building 3 31 9.1 85.3 98.1 Commercial Building 4 71.4 43 3.9 6.5 Commercial Building 5 58.6 26.1 11.59 42.6 Commercial Building 6 0 0 0 0 Hotel Building 7 58 69.6 92.6 65.1 Office Building 8 14.2 0.1 96.2 100 Residential Building 1 0 0 0 0 Residential Building 2 57.5 7.6 41.6 47.5 Residential Building 3 92.5 100 98.4 98.3 Residential Building 4 34.1 16.4 57.5 45.6 Residential Building 5 0 0 0 0 SONNE LEC Aggregate 97.6 56.5 80.7 96.9 5.2.3.2. Energy Cost and Net Energy Traded of SONNE LEC v.3 With the introduction of P2P trading in the SONNE community, savings were calculated for each building and the community as a whole, both in absolute and percentage values compared to the base case (see Tables 8a and 8b below for the two simulated periods). For the simulated week in the summer period (1-7 August 2023), the community generated a profit of €45.1, in contrast to the base case scenario v.1, where the community incurred an aggregate net cost of €100. This represents a 145.1% increase in monetary benefits. This result affirms the advantage of P2P trading, which enables energy consuming and producing assets in the community to access competitive prices, leading to important economic benefits. November simulation results did not exhibit such a dramatic benefit, due to the lower renewable energy production of the community during that time period and higher demand, reducing the scope for trading. Nevertheless, there was still a 10.1% increase in the monetary benefits for the community with activated P2P trading in this winter period compared to the base case scenario, since the community accrued an aggregate net cost of €755.9, instead of €840.5. Importantly, all community participants benefited, with those with higher flexibility accruing a larger share of the benefits, as shown in the last column of the two tables. D3.5 Description and report of energy exchange and smart trading tool including simulation studies for living labs as active LECs. Preliminary version 66 Table 8a: Electricity and heating cost, revenue and balance for SONNE LEC and per participant, simulation v.3 for 1-7 August 2023, with v.1 benefits comparison Building ID Electricity Cost (€) Heating Cost (€) Revenue (€) Balance (€) Benefits compared to Base Case v.1 (€) Benefits compared to Base Case v.1 (%) Commercial Building 1 4.3 3.7 24.1 -16.1 2.4 17.4 Commercial Building 2 8.1 8.3 0 16.4 0.8 4.4 Commercial Building 3 76.3 8.4 5.4 79.3 6.9 8 Commercial Building 4 861.8 37.2 1306.2 -407.1 49.9 14 Commercial Building 5 254.3 - 479.7 -225.4 20.1 9.8 Commercial Building 6 21.7 - 0 21.7 2.1 8.7 Hotel Building 7 492 33.1 392.1 132.9 16.5 11 Office Building 8 206.8 8.6 1.2 214.2 21.3 9 Residential Building 1 48.6 - 0 48.6 4.5 8.5 Residential Building 2 15.2 - 26.8 -11.6 3.5 43 Residential Building 3 910.7 - 884.4 26.4 4.3 14.1 Residential Building 4 111.5 - 39.3 72.2 12.5 14.7 Residential Building 5 3.4 - 0 3.4 0.4 9.3 SONNE LEC Aggregate 3014.8 99.2 3159.1 -45.1 145.1 145.1 D3.5 Description and report of energy exchange and smart trading tool including simulation studies for living labs as active LECs. Preliminary version 67 Table 8b. Electricity and heating cost, revenue and balance for SONNE LEC and per participant, simulation v.3 for 1-7 November 2023, with v.1 benefits comparison Building ID Electricity Cost (€) Heating Cost (€) Revenue (€) Balance (€) Benefits compared to Base Case v.1 (€) Benefits compared to Base Case v.1 (%) Commercial Building 1 7.9 22.4 7.6 22.7 1 4.3 Commercial Building 2 10.2 31.4 0 41.6 0.5 1.1 Commercial Building 3 141.3 41.9 0.2 183 5.9 3.1 Commercial Building 4 402.5 88.3 576.9 -86.1 40.1 87.4 Commercial Building 5 258 - 169 89 8.3 8.5 Commercial Building 6 23.8 - 0 23.8 1.1 4.6 Hotel Building 7 228.3 32.2 201.5 59 3.7 5.9 Office Building 8 125.4 36.8 0 162.2 5.4 3.2 Residential Building 1 58.2 - 0 58.2 2.6 4.3 Residential Building 2 73.2 - 6 67.2 2.5 3.5 Residential Building 3 419 - 407.5 11.5 5.5 32.3 Residential Building 4 150.4 - 31.6 118.8 7.8 6.1 Residential Building 5 4.9 - 0 4.9 0.2 4.2 SONNE LEC Aggregate 1903.2 253 1400.4 755.9 84.6 10.1 5.2.3.3. Energy Import and Export of SONNE LEC v.3 A graph of the energy imports and exports in scenario v.3 is shown in Figures 31a and 31b below for the two simulated periods. The energy imports and exports are following the same trend as the measured self-sufficiency and self-consumption values that were reported above. In comparison to scenario v1, we can see a significant decrease in the energy imports (from 2754.6 kWh to 211.4 D3.5 Description and report of energy exchange and smart trading tool including simulation studies for living labs as active LECs. Preliminary version 68 kWh), amounting to 92.32% for August and 36.60% for November results (from 3202.8 kWh to 2030.5 kWh). We also observe an important decrease in exports from the community to the grid, amounting to 71.75% for August (from 3508.1 kWh to 991.1) and 93.50% for November results (from 1146.2 kWh to 74.56 kWh), and conversely a significant increase in the energy imports and exports between community buildings. Figure 31a. Net energy import and export graph for the SONNE LEC, simulation v.3 for 1-7 August 2023 with the horizontal axis showing the building IDs and the vertical axis showing exported and imported energy in kWh Figure 31b. Net energy import and export graph for the SONNE LEC, simulation v.3 for 1-7 November 2023 with the horizontal axis showing the building IDs and the vertical axis showing exported and imported energy in kWh 5.2.4. Activated P2P trading, with heat pumps replacing district heating (SONNE LEC v.4) In SONNE LEC v.4 scenario, the current conditions of SONNE are reflected in terms of available assets (modelled based on provided historical data and supplementary data as described above), with one modification: “virtual heat pumps” with storage are introduced to all buildings in the community that had a district heating connection, simulating their replacement with heat pumps to satisfy the heat demand (hence the scenario assumes no district heating connection). The current market conditions in terms of utility and feed-in-tariff rate apply, but the market mechanism is altered to pay-as-bid peer-to-peer (P2P) trading, enabling local energy market (LEM) participants (community members) to trade surplus (in this case PV-based) electricity generation among each other. D3.5 Description and report of energy exchange and smart trading tool including simulation studies for living labs as active LECs. Preliminary version 69 5.2.4.1. Self-Consumption and Self-Sufficiency of SONNE LEC v.4 With heat pumps satisfying the heat demand instead of the district heating network and activated P2P trading, the self-sufficiency levels of the community increased from 39.6% to 87.8% in August, and from 21.7% to 37.5% in November, as compared to the scenario v.2 where VHPs replaced district heating but there was no P2P trading. Additionally the self-consumption values increased from 42.1% to 87.5% in August, and from 58.0% to 100% in November. These results further confirm the multiplication of renewable and especially flexibility asset benefits by enabling P2P trading. The heat pump flexibility is responsible for 100% self-consumption achieved in the winter period. The self-sufficiency and self-consumption rates for the SONNE LEC v.4 scenario for the first week in August and November 2023, respectively, are shown in percentage terms in Table 9 below. The self-sufficiency and self-consumption levels of most of the individual buildings remained as in scenario v.2 since the energy production and energy demand of these buildings was constant. However in commercial buildings 4 and 5, the self-sufficiency rate exhibited a stark decrease in the August results, from 62.5% to 52.1% (16.6% decrease) and 84.5% to 58.4% (30.9% decrease), respectively. This is due to the fact that the electricity prices in the community are higher, incentivising buildings 4 and 5 to sell their own PV production for a profit rather than use it for self-consumption. For instance, building 4 has an export of PV energy of 5459.2 kWh to other buildings for this period, which approximately reduces its self sufficiency by 10.4%. Table 9. Self-sufficiency and self-consumption rates for SONNE LEC and per participant, simulation v.4 for 1-7 August/November 2023 Building ID Self-Sufficiency August (%) Self-Sufficiency November (%) Self-Consumpti on August (%) Self-Consumpti on November (%) Commercial Building 1 32.6 8.9 23.7 40.5 Commercial Building 2 0 0 0 0 Commercial Building 3 26 5.7 85.3 88.9 Commercial Building 4 52.1 20.1 10.8 22.5 Commercial Building 5 58.4 25.8 11.4 38.3 Commercial Building 6 0 0 0 0 Hotel Building 7 46.4 37.1 93.1 60.7 Office Building 8 13 0.1 96.5 81.5 Residential Building 1 0 0 0 0 Residential Building 2 57.5 7.6 41.6 38.5 Residential Building 3 92.5 100 98.4 84.8 D3.5 Description and report of energy exchange and smart trading tool including simulation studies for living labs as active LECs. Preliminary version 70 Residential Building 4 34.1 16.4 57.5 39.9 Residential Building 5 0 0 0 0 SONNE LEC Aggregate 87.8 37.5 87.5 100 If we compare this scenario to scenario v.3 when P2P was also activated, we can deduce that replacing district heating by heat pumps in an actively trading energy community would increase the self-sufficiency level of the community slightly - from 83.1% to 87.8% in August, and from 56.5% to 37.5% in November. The reason for such a small difference lies in insufficient local electricity generation to cover the additional consumption required for heat pump operation. This also explains the rise in the self-consumption rate, from 80.7% to 87.5% in August, and from 96.9% to 100% in November. The behaviour of the battery is altered in scenario v.4 when compared to scenario v.3 as a consequence of the heat pump buying strategy, prioritising own consumption. For instance, in scenario v.4, there is increased energy demand in the Commercial Building 4 due to the introduction of virtual heat pumps. On the first day of the simulation, these virtual heat pumps draw energy from the battery of Commercial Building 4, as indicated by the state of charge (SOC) of the battery on the first day of August in Fig. 32 below. We observe that the peak SOC is less than 100%, in contrast to other days of that week that peak at 100%. The heat pumps self-consume from the battery to reduce reliance for operation both from the rest of the community and the external grid. At the same time, the PV of Commercial Building 4 exports its production to the community maximising the monetary benefit of the building. Figure 32. Battery state-of-charge (green), traded energy in kWh (purple) and energy rate in Euro/kWh (blue), Commercial Building 4 in SONNE LEC, simulation v.4 for 1-7 August 2023 with the horizontal axis showing timestamps for the selected week and the vertical axis showing the aforementioned variables D3.5 Description and report of energy exchange and smart trading tool including simulation studies for living labs as active LECs. Preliminary version 71 On the other hand, the SOC of the battery in the Commercial Building 4 in scenario v.3, has a peak of 100% each day of the week indicating that the battery sells its energy during times of no PV production at night and fully charges during the day when PV production is high. Moreover, in comparison to scenario v.3, we observe that on average the SoC of the battery is higher compared to scenario v.4 due to the shorter timeframe that the storage is empty, which indicates that the battery gets depleted slower due to the decreased electricity demand. This behaviour is as expected as the building has a high PV production and does not have virtual heat pumps. Figure 33. Battery state-of-charge (green), traded energy in kWh (purple) and energy rate in Euro/kWh (blue), Commercial Building 4 in SONNE LEC, simulation v.3 for 1-7 August 2023 with the horizontal axis showing timestamps for the selected week and the vertical axis showing the aforementioned variables 5.2.4.2. Energy Cost and Net Energy Traded of SONNE LEC v.4 With the introduction of both P2P trading district and heating replacement by heat pumps in the SONNE community, the community in aggregate and all buildings in the community individually achieve savings in the simulated week in August (see Table 10a), with the Commercial Building 4 enjoying the largest benefit of about €375.1, which represents a 17.8% improvement compared to scenario v.2. The community as a whole has an aggregate net cost of €72.8 compared to the base case scenario v.2., where that cost is €234.3, which represents 68.9% improvement. If we compare the results to the base case scenario then the absolute benefit is €31,1, and the percentage benefit is 31.1%. We observe similar outcomes for the November results (see Table 10b), albeit less pronounced due to the decreased renewable energy production of the community in comparison to August (6.5% improvement, with net cost reduced from €1150.5 to €1075.4) when comparing scenarios v.4 and v.2. However, when we compare v.4 with the base case scenario v.1, we see that the net cost is still higher in winter months if heat pumps replace district heating, even when P2P is introduced (€1075.4 vs. €840.5). These results indicate that it is difficult to overcome the low cost of district heating, and that heat pumps could potentially optimally replace district heating if peer-to-peer trading is enabled at a larger scale and if the community invested in alternative resources to PV, including storage to ensure higher renewable capacity in the winter period. D3.5 Description and report of energy exchange and smart trading tool including simulation studies for living labs as active LECs. Preliminary version 72 Table 10a. Electricity cost, revenue, balance for SONNE LEC and per participant, simulation v.4 for 1-7 August 2023, with v.2 benefits comparison Building ID Energy Cost (€) Revenue (€) Balance (€) Benefits compared to v.2 (€) Benefits compared to v.2 (%) Commercial Building 1 17 24.1 -7 2.4 52.7 Commercial Building 2 30.6 0 30.6 1.7 5.2 Commercial Building 3 99.3 5.5 93.8 7.3 7.3 Commercial Building 4 857.6 1232.7 -375.1 56.7 17.8 Commercial Building 5 243.9 472.3 -228.4 23.2 11.3 Commercial Building 6 21.9 0 21.9 1.9 7.9 Hotel Building 7 540.6 370.2 170.3 20 10.5 Office Building 8 230.7 1.2 229.5 21.1 8.4 Residential Building 1 48.9 0 48.9 4.3 8 Residential Building 2 15.3 27.3 -12 3.9 48 Residential Building 3 847.1 822.8 24.3 6.4 20.8 Residential Building 4 112.6 40.1 72.5 12.2 14.4 Residential Building 5 3.5 0 3.5 0.3 8.9 SONNE LEC Aggregate 3069.1 2996.2 72.8 161.4 68.9 D3.5 Description and report of energy exchange and smart trading tool including simulation studies for living labs as active LECs. Preliminary version 73 Commercial Building 5 52.7 55.7 -3 100.2 103.1 Commercial Building 6 10.3 0 10.3 14.6 58.5 Hotel Building 7 24.6 22.3 34.5 28.2 45 Office Building 8 55.3 0 92.1 75.6 45.1 Residential Building 1 23.5 0 23.5 37.3 61.3 Residential Building 2 32.2 5.4 26.8 42.8 61.5 Residential Building 3 25.1 0.9 24.2 -7.3 -43 Residential Building 4 48.7 28.6 20 106.6 84.2 Residential Building 5 1 0 1 4.2 81.2 SONNE LEC Aggregate 361.3 291.4 322.9 517.6 61.6 5.2.5.2.2 Day ToU Results of SONNE LEC v.5 With the introduction of day ToU tariffs in the SONNE community, savings were calculated for each building and the community as a whole, both in absolute and percentage values compared to the base case. In the August results (see Table 13a), the community incurred an aggregate net cost €8.4, in contrast to the base case scenario v.1, where the community incurred an aggregate net cost of €100 and to scenario v.3, where the community generated a profit of €45.1. Proportionally, this represents a 91.6% increase in monetary benefits in comparison to scenario v.1 and a 81.4% decrease in monetary benefits in comparison to scenario v.3, which stands in stark contrast to Night ToU results presented above. The outcome is attributed to the decreased utility electricity rates during the day, which coincides with the PV production in the community. The decreased utility rates incentivise community consumers to buy energy from the utility and not self-consume, thus self-consuming less overall. The revenue generated from the PVs is significantly higher compared to the night ToU tariff (€2314.6 contrary to €759.6), but the increase is insufficient to offset the higher prices that the community consumers would be obliged to pay overall. The conclusion from this scenario simulation is that the ToU tariffs are economically beneficial for the consumers only when they do not coincide with the times of high self-consumption. D3.5 Description and report of energy exchange and smart trading tool including simulation studies for living labs as active LECs. Preliminary version 80 Table 13a. Electricity cost, revenue and balance for SONNE LEC and per participant, simulation v.5 with day ToU tariff for 1-7 August 2023, with v.1 benefits comparison Building ID Energy Cost (€) Revenue (€) Balance (€) Benefits compared to Base Case v.1 (€) Benefits compared to Base Case v.1 (%) Commercial Building 1 7.7 25.7 -14.3 0.6 4.2 Commercial Building 2 7.9 0 16.1 1 5.8 Commercial Building 3 93 23.1 78.3 7.9 9.2 Commercial Building 4 518.4 926.5 -371 13.7 3.8 Commercial Building 5 161.8 372.1 -210.3 5.1 2.5 Commercial Building 6 21 0 21 2.7 11.5 Hotel Building 7 411.7 307.9 136.9 12.6 8.4 Office Building 8 217.4 20.4 205.6 29.8 12.7 Residential Building 1 46.9 0 46.9 6.3 11.8 Residential Building 2 24.6 34.6 -10 1.9 22.8 Residential Building 3 571.2 538.8 32.4 -1.7 -5.7 Residential Building 4 138.9 65.6 73.3 11.4 13.5 Residential Building 5 3.3 0 3.3 0.5 13 SONNE LEC Aggregate 2223.8 2314.6 8.4 91.6 91.6 The increase of the monetary benefits in comparison to scenario v.1 is also evident in the November results (see Table 13b), only less pronounced due to the lower renewable production in the winter period. The community accrued an aggregate net cost of €755.6, in contrast to the base case scenario v.1, where the community incurred an aggregate net cost of €840.5, representing a 10.1% increase in monetary benefits. Contrary to that, the monetary benefits were only marginally increased compared D3.5 Description and report of energy exchange and smart trading tool including simulation studies for living labs as active LECs. Preliminary version 81 to scenario v.3, when this value was €755.9. The outcome of the analysis is that applying Day ToU tariffs does not provide a financial or an environmental incentive to the community members, since their total cost is similar to the P2P energy trading scenario v.3. Table 13b. Electricity cost, revenue and balance for SONNE LEC and per participant. simulation v.5 with day ToU tariff for 1-7 November 2023, with v.1 benefits comparison Building ID Energy Cost (€) Revenue (€) Balance (€) Benefits compared to Base Case v.1 (€) Benefits compared to Base Case v.1 (%) Commercial Building 1 11.2 10.5 23.2 0.6 2.5 Commercial Building 2 9.6 0 41 1.1 2.7 Commercial Building 3 140.1 10 172.1 16.8 8.9 Commercial Building 4 572.3 714.9 -54.4 8.5 18.5 Commercial Building 5 304.9 218.1 86.8 10.4 10.7 Commercial Building 6 22.2 0 22.2 2.8 11.1 Hotel Building 7 297.4 268.4 61.1 1.5 2.5 Office Building 8 116.8 0.1 153.5 14.1 8.4 Residential Building 1 54.4 0 54.4 6.5 10.7 Residential Building 2 72.4 9.4 63 6.7 9.6 Residential Building 3 564.7 549 15.7 1.3 7.5 Residential Building 4 159.4 47 112.5 14.1 11.1 Residential Building 5 4.6 0 4.6 0.5 9.8 SONNE LEC Aggregate 2330 1827.4 755.6 84.9 10.1 D3.5 Description and report of energy exchange and smart trading tool including simulation studies for living labs as active LECs. Preliminary version 82 The Day ToU has a low market maker rate (external utility rate) during the day when the PV production is high, which incentives the batteries to buy both cheaper PV and market maker (utility grid) electricity during the day and sell excess in the evening at a higher rate making a profit. Fig. 35 below illustrates how the battery charges quickly with the readily available, low priced electricity during the day, and then slowly discharges in the evening and night-time as it is selling electricity at a higher rate. Figure 35. Battery state-of-charge (green), traded energy in kWh (purple) and energy rate in Euro/kWh (blue), Commercial Building 4 in SONNE LEC, simulation v.5 for 1-7 August 2023 with the horizontal axis showing timestamps for the selected week and the vertical axis showing the aforementioned variables 5.2.5.3. Energy Import and Export of SONNE LEC v.5 A graph of the energy imports and exports in the SONNE scenario v.5 is shown in Figures 36 a-d below for the two simulated periods and two applied tariff models. As inferred from the self-sufficiency and self-consumption analysis, In August, the energy exported with Day ToU tariffs increased by 80.85% compared to the Night ToU scenario (from 2330.4 kWh to 4214.5 kWh), and the energy imported increased by 284.72% (from 913.10 kWh to 3512.84 kWh). In November, the energy exported with Day ToU tariffs increased by 150.79% compared to the Night ToU scenario (from 820.1kWh to 2056.8kWh), and the energy imported increased by 268.15% (from 1133.3 kWh to 4172.2 kWh). With the utility offering more affordable energy prices during the day, when the renewable production from the PVs is high, the community is incentivized to import energy from the grid, and not self-consume but sell the energy generated by its PVs to the grid to obtain a better revenue. When the Night ToU tariffs are implemented we observe a significant decrease in the energy exports and imports compared to the scenario with Day ToU tariffs. The reason for this is that now the utility offers more affordable energy rates during the night, when there is no renewable production from the PVs in the community. Therefore, the energy produced by the PVs is maximally self-consumed, including to charge the batteries, hence decreasing the amount of energy traded with the grid. D3.5 Description and report of energy exchange and smart trading tool including simulation studies for living labs as active LECs. Preliminary version 83 We observe very similar results in the imported and exported energy levels when comparing scenario v.3 with the Night ToU tariff application in scenario v.5, since this tariff scheme is closer to the pricing pattern of scenario v.3. The imported energy is 331.77% higher in the August period (from 211.44 kWh to 913.10 kWh) and 44.19% lower in the November period with Night ToU tariffs due to the utility rate decreasing after 14h, which incentivises a higher acquisition of electricity from the utility. When comparing scenario v.3 with the Day ToU tariff application in scenario v.5, we observe a 1561.39% increase in the imported energy (from 211.44 kWh to 3512.84 kWh) and 325.23% increase in energy exports (from 991.11 kWh to 4214.5 kWh). November results follow a similar trend, with a 105.48% increase in energy imports (from 2030.47 kWh to 4172.2 kWh) and 2658.62% increase in energy exports (74.56 kWh to 2056.8 kWh). This is attributed to the inexpensive utility rates during the time that PVs are operating in the Day ToU tariff model (similar to the comparison with the night ToU scenario). Figure 36a. Net energy import and export graph for the SONNE LEC, simulation v.5 with Day ToU tariffs, for 1-7 August 2023 with the horizontal axis showing the building IDs and the vertical axis showing the exported and imported energy in kWh Figure 36b. Net energy import and export graph for the SONNE LEC, simulation v.5 with Day ToU tariffs for 1-7 November 2023 with the horizontal axis showing the building IDs and the vertical axis showing exported and imported energy in kWh D3.5 Description and report of energy exchange and smart trading tool including simulation studies for living labs as active LECs. Preliminary version 84 Figure 36c. Net energy import and export graph for the SONNE LEC, simulation v.5 with Night ToU tariffs for 1-7 August 2023 with the horizontal axis showing the building IDs and the vertical axis showing exported and imported energy in kWh Figure 36d. Net energy import and export graph for the SONNE LEC, simulation v.5 with Night ToU tariffs, for 1-7 November 2023 with the horizontal axis showing the building IDs and the vertical axis showing exported and imported energy in kWh 5.2.6. Activated P2P trading with applied dynamic tariffs and heat pumps replacing district heating (SONNE LEC v.6) In SONNE LEC v.6 scenario, the current conditions of SONNE are reflected in terms of available assets (modelled based on provided historical data and supplementary data as described above), with one modification: “virtual heat pumps” with storage are introduced to all buildings in the community that had a district heating connection, effectively replacing district heating by heat pumps to satisfy the heat demand (hence the scenario assumes no district heating connection). The current market conditions apply in terms of utility and feed-in-tariff rates, but the market mechanism is altered to pay-as-bid peer-to-peer (P2P) trading, enabling LEM participants (community members) to trade surplus (in this case PV-based) electricity generation among each other. In addition, two time-of-use (ToU) tariffs are introduced. 5.2.6.1. Self-Consumption and Self-Sufficiency of SONNE LEC v.6 The self-sufficiency and self-consumption rates for the SONNE LEC v.6 scenario for the first week in August and November 2023, respectively, are shown in percentage terms in Table 14 below and analysed in comparison to scenario v.2 which also had virtual heat pumps modelled to simulate D3.5 Description and report of energy exchange and smart trading tool including simulation studies for living labs as active LECs. Preliminary version 85 replacement of district heating but no P2P trading or dynamic tariffs. As in the previous, v.5 scenario, results differ substantially depending on the deployed tariff scheme. With the Day ToU tariff model (utility rates reduced during day-time when PV generation is peaking, disincentivizing self-consumption), the self-sufficiency of the community significantly decreased compared to scenario v.2 in the summer period (from 39.6% to 28.7%), and even further in the winter period (from 21.7% to 5.3%). The same observation applies to the self-consumption rates, which decreased from 42.1% to 28.5% in August, and from 58% to 14.2% in November, compared to scenario v.2. In contrast, when Night ToU tariffs were deployed in the community (utility rates more affordable during night-time, incentivising self-consumption during the day when PV generation is high), the self-sufficiency of the community significantly increased compared to scenario v.2, from 39.6% to 65.2% in August, and from 21.7% to 33.4% in November. Similarly, the self-consumption increased from 42.1% to 64.9% in August, and from 58% to 89.4% in November. The flexibility provided by the storage and the (virtual) heat pumps plays a significant role in the self-sufficiency and self-consumption increase, since the PV production is stored to be used later, further decreasing the need for consuming energy from the grid. Table 14. Self-sufficiency and self-consumption rates for SONNE LEC and per participant for day and night time-of-use tariffs in simulation v.6 for 1-7 August 2023 Building ID Day ToU Night ToU SelfSufficiency August (%) SelfSufficiency November (%) SelfConsumption August (%) SelfConsumption November (%) SelfSufficiency August(%) SelfSufficiency November (%) SelfConsumption August (%) SelfConsumption November (%) Commercial Building 1 15.8 1.4 10.4 6.2 26.7 4.2 18.5 18.5 Commercial Building 2 0 0 0 0 0 0 0 0 Commercial Building 3 10 0.9 29.5 13.6 21.7 2.9 68.7 55 Commercial Building 4 25.1 3.7 5.1 4.1 26.2 15.3 5.8 14.7 Commercial Building 5 32.2 6.8 5.8 9.7 52.6 18.1 10.3 31 Commercial Building 6 0 0 0 0 0 0 0 0 Hotel Building 7 15.5 9.3 30.5 15.1 43.7 24.2 88.8 46.6 Office Building 8 4.6 0 32.4 28.1 10.8 0 78.3 16.2 Residential Building 1 0 0 0 0 0 0 0 0 Residential 26.8 1.2 17.5 5.3 47 3.4 34.1 19.2 D3.5 Description and report of energy exchange and smart trading tool including simulation studies for living labs as active LECs. Preliminary version 86 Building 2 Residential Building 3 31.2 15.9 30.5 7.4 83.2 96.7 87.8 66.2 Residential Building 4 14.3 3.7 22.6 8.1 30.1 10.4 51.1 29.1 Residential Building 5 0 0 0 0 0 0 0 0 SONNE LEC Aggregate 28.7 5.3 28.5 14.2 65.2 33.4 64.9 89.4 5.2.6.2. Energy Cost and Net Energy Traded of SONNE LEC v.6 With the introduction of dynamic tariffs (ToU) and P2P trading in the SONNE community, the community in total and all buildings in the community derive absolute monetary savings. Similarly to scenario v.5, different outcomes are observed depending on the implemented ToU tariff model. 5.2.6.2.1 Night ToU Results of SONNE LEC v.6 With the introduction of Night ToU tariffs in the SONNE community, savings were calculated for each building and the community as a whole, and then compared to scenarios v.2 and v.4 both in absolute and percentage terms, as presented in Table 15a below for the simulated week in August and Table 15b for the simulated week in November. In the summer period, the community generated a profit of €97.8, in contrast to scenario v.2, where the community incurred an aggregate net cost of €234.3 and to scenario v.4, where the community generated a profit of €72.8. This represents a 141.7% increase in monetary benefits in comparison to scenario v.2. Table 15a. Electricity cost, revenue and balance for SONNE LEC and per participant, simulation v.6 with Night ToU tariff for 1-7 August 2023, with v.2 benefits comparison Building ID Energy Cost (€) Revenue (€) Balance (€) Benefits compared to scenario v.2 (€) Benefits compared to scenario v.2 (%) Benefits compared to scenario v.4 (€) Benefits compared to scenario v.4 (%) Commercial Building 1 14.9 23.8 -8.9 4.3 92.6 1.9 27.1 Commercial Building 2 26.6 0 26.6 5.7 17.8 4.0 13.1 Commercial Building 3 70.8 10.7 60.1 41 40.5 33.7 35.9 Commercial Building 4 184.8 512 -327.2 8.8 2.8 -47.9 -12.8 Commercial 56 275.8 -219.8 14.6 7.1 -8.6 -3.8 D3.5 Description and report of energy exchange and smart trading tool including simulation studies for living labs as active LECs. Preliminary version 87 Building 5 Commercial Building 6 14.2 0 14.2 9.5 40 7.7 35.2 Hotel Building 7 198.7 77.2 121.6 68.8 36.2 48.7 28.6 Office Building 8 175.5 6.8 168.7 81.9 32.7 60.8 26.5 Residential Building 1 31.4 0 31.4 21.7 40.9 17.5 35.8 Residential Building 2 9.3 29.1 -19.8 11.7 143.5 7.8 65.0 Residential Building 3 155.4 123.6 31.9 -1.2 -3.8 -7.6 -31.3 Residential Building 4 64.5 43.5 21 63.8 75.3 51.5 71.0 Residential Building 5 2.3 0 2.3 1.5 39.4 1.2 34.3 SONNE LEC Aggregate 1004.6 1102.4 -97.8 332.1 141.8 170.6 234.3 Similar outcomes are observed for the winter period, only less pronounced again due to the lower renewable energy production. The community accrued an aggregate net cost of €487.9, in contrast to scenario v.2, where the community incurred an aggregate net cost of €1150.5, and to scenario v.4 (€1075.4). This represents a 57.6% increase in monetary benefits for the community in comparison to scenario v.2. Table 15b. Electricity cost, revenue and balance for SONNE LEC and per participant, simulation v.6 for 1-7 November 2023 with Night ToU tariff , with v.2 benefits comparison Building ID Energy Cost (€) Revenue (€) Balance (€) Benefits compared to scenario v.2 (€) Benefits compared to scenario v.2 (%) Benefits compared to scenario v.4 (%) Commercial Building 1 53.6 9.9 43.7 10.5 19.4 17.9 Commercial Building 2 71.9 0 71.9 12.1 14.4 12.2 Commercial Building 3 140.5 6.6 133.9 110.2 45.1 43.6 Commercial Building 4 615.7 598.2 17.5 31 63.9 -29.6 Commercial Building 5 181.2 202.3 -21.1 118.4 121.7 123.3 Commercial Building 6 9 0 9 15.9 63.8 62.7 D3.5 Description and report of energy exchange and smart trading tool including simulation studies for living labs as active LECs. Preliminary version 88 Hotel Building 7 293.4 242.1 51.2 52.3 50.5 48.9 Office Building 8 128.1 0.1 128 86.7 40.4 38.6 Residential Building 1 20 0 20 40.8 67.1 66.2 Residential Building 2 30.6 8.6 22 47.7 68.4 67.6 Residential Building 3 478.9 476 2.9 14 82.8 78.7 Residential Building 4 46.5 38.6 7.9 118.7 93.8 93.4 Residential Building 5 0.9 0 0.9 4.3 83.4 82.0 SONNE LEC Aggregate 2070.3 1582.5 487.9 662.6 57.6 54.6 5.2.6.2.2 Day ToU Results of SONNE LEC v.6 With the introduction of Day ToU tariffs in the SONNE community, , savings were calculated for each building and the community as a whole, and then compared to scenarios v.2 and v.4. In the summer period (see Table 16a), the community incurred an aggregate net cost of €119.1, in contrast to scenario v.2, where it incurred an aggregate net cost of €234.3 and to scenario v.4, where the community generated a profit of €72.8. This represents a 49.2% increase in monetary benefits in comparison to scenario v.2. For the same reasons as in scenario v.5, the economic results for the community were much worse than with the Night ToU tariff model implementation. This further contributes to the conclusion that ToU tariffs have to be carefully curated in order to engage prosumers and flexible consumers to support grid resilience. Table 16a. Electricity cost, revenue and balance for SONNE LEC and per participant, simulation v.6 for 1-7 August 2023 with Day ToU tariff, with v.2 benefits comparison Building ID Energy Cost (€) Revenue (€) Balance (€) Benefits compared to scenario v.2 (€) Benefits compared to scenario v.2 (%) Benefits compared to scenario v.4 (%) Commercial Building 1 20.6 25.6 -5 0.4 8.6 -28.6 Commercial Building 2 30.4 0 30.4 2 6.1 0.7 Commercial Building 3 115.1 23.1 92 9.1 9 1.9 Commercial Building 4 786.6 1126.9 -340.3 21.9 6.9 -9.3 Commercial Building 5 220.8 433.9 -213.1 7.9 3.8 -6.7 Commercial 21 0 21 2.7 11.4 4.1 D3.5 Description and report of energy exchange and smart trading tool including simulation studies for living labs as active LECs. Preliminary version 89 Residential Building 2 15.3 27.5 -12.2 4.1 50.1 1.7 Residential Building 3 898.3 873.7 24.6 6.1 20 -1.2 Residential Building 4 113.2 40.4 72.8 11.9 14.1 -0.4 Residential Building 5 3.5 0 3.5 0.3 8.5 0.0 SONNE LEC Aggregate 3109.7 3050.7 59 175.3 74.8 19.0 Table 18b. Electricity cost, revenue and balance for SONNE LEC and per participant, simulation v.7 for 1-7 November 2023, with v.2 benefits comparison Building ID Energy Cost (€) Revenue (€) Balance (€) Benefits compared to scenario v.2 (€) Benefits compared to scenario v.2 (%) Benefits compared to scenario v.4 (%) Commercial Building 1 54.7 5.9 48.8 5.4 9.9 8.3 Commercial Building 2 74.4 0 74.4 9.6 11.5 9.2 Commercial Building 3 228.8 0.2 228.6 15.4 6.3 3.7 Commercial Building 4 187.9 202.2 -14.3 62.9 129.5 205.9 Commercial Building 5 168.7 76.4 92.3 4.9 5 -2.0 Commercial Building 6 24.6 0 24.6 0.3 1.3 -2.1 Hotel Building 7 125.9 32 93.9 9.6 9.3 6.3 Office Building 8 199.2 0 199.2 15.5 7.2 4.5 Residential Building 1 60.1 0 60.1 0.7 1.2 -1.7 Residential Building 2 74.8 6.1 68.8 0.9 1.3 -1.5 Residential Building 3 53.3 37.9 15.4 1.6 9.4 -13.2 Residential Building 4 154.1 32.2 121.9 4.7 3.7 -1.2 Residential Building 5 5.1 0 5.1 0 0.5 -2.0 SONNE LEC Aggregate 1411.6 392.8 1018.8 131.7 11.4 5.3 5.2.7.3. Energy Import and Export of SONNE LEC v.7 A graph of the energy imports and exports in the SONNE scenario v.7 is shown in Figures 39a and 39b below for the two simulated periods. In comparison to scenario v.4, the November results are almost identical, which is attributed to the fact that there is no excess renewable energy from the PVs that could be used for flexibility. The heat and electricity demand during winter periods are significantly D3.5 Description and report of energy exchange and smart trading tool including simulation studies for living labs as active LECs. Preliminary version 96 higher than the renewable PV production in the community, which has diminishing returns on the added flexibility based on smart trading. Contrary to the November results, in the August results we see a considerable decrease in the imports and exports from the grid in comparison to scenario v.4. This indicates that the extra flexibility harnessed by the smart trading algorithm was very beneficial for the community, with the virtual heat pumps avoiding to buy energy from the utility when the prices were not affordable. Therefore, we deduce that the smart trading algorithm has both economic and environmental benefits, incentivising local consumption of local generation and maximising flexibility provided by heat pump storages by following price signals. Figure 39a. Net energy import and export graph for the SONNE LEC, simulation v.7 for 1-7 August 2023 with the horizontal axis showing the building IDs and the vertical axis showing exported and imported energy in kWh Figure 39b. Net energy import and export graph for the SONNE LEC, simulation v.7 for 1-7 November 2023 with the horizontal axis showing the building IDs and the vertical axis showing exported and imported energy in kWh D3.5 Description and report of energy exchange and smart trading tool including simulation studies for living labs as active LECs. Preliminary version 97 6 Conclusions The activities outlined in this report, principally based on the HYPERGRYD project WP3 - ICT Modules and Simulation Tools T3.4: Energy exchange and smart trading tool for LEC with coupled networks [1], fulfil this task’s primary objectives: ●Provide decision support and trading simulation (digital twins) to the grid operators and end users for distributed energy exchange in LEC with coupled DHC and electric grid (such as peer-to-peer, exporting to the grid, etc.) ●Develop GSY open-source energy exchange engine to create features to digitally represent distributed heat pumps and district heating and integrate these features to simulate a local grid-aware energy marketplace, where peer-to-peer transactions are enabled among distributed energy assets based on a hierarchical market topography ●Develop Proof-of-Concept LEC simulations in a regulatory sandbox environment (participating living labs), integrating enhanced grid management (improved tariff design) and energy asset aggregators (improved, smart trading algorithms) and engaging relevant regulators to consider wider implementation. As elaborated in the Interim Project Report [1], dynamic model tariff testing in collaboration with grid operators was not possible because the local grid operators were not interested in active project participation. Nonetheless, the demonstration study included two different dynamic tariff models that could serve to improve the local grid operators and the regulators about the potential of these regulation tools in flexibility and congestion management. Task 3.4, as envisaged, fed on KPIs and regulatory and market aspects defined in T1.4 and used the SONNE data. KTH supported mathematical model development and selected algorithm embedding (from Task 3.2) for virtual heat pump modelling. Finally GSY proposed and tested a new trading algorithm to improve energy asset (heat pump) trading strategies in LEC. This work has also relied on the ICT architecture and use case research and development tasks in WP4 - HYPERGRYD Digital Twin Platform as a service, and tasks performed in WP5 -TRL5 demonstration in living labs and virtual labs in LEC, especially Task 5.5 Model demonstration and validation of the HYPERGRYD ICT services at SONNE. Notably, this report contributes to the following WP5 objectives: ●demonstrating and validating the modelling tool developed in Task 3.5 by applying it to specific models of the LEC networks developed within the scope of this project, ●assessing the environmental, economic and social impacts of the HYPERGRYD solutions, environmental study and deriving the impacts of the applied technology. Validation testing results, both in terms of measurement validation and trading results analysis, align with the expectations, confirming specifically that the current implementation of the heat pump in D3.5 Description and report of energy exchange and smart trading tool including simulation studies for living labs as active LECs. Preliminary version 98 the Grid Singularity Exchange [2,3] is functioning as intended. The heat pump's trading strategies adhere to the provided consumption profiles, and the storage temperature follows the pattern of purchased energy. The maximum power rating is also adhered to for all heat pumps, resulting in the expected unmatched demand in time slots where the demanded consumption exceeded this value. Finally, the heat pump software testing informed future use of the GSY Simulation Tool, indicating that the HP digital twin’s trading strategy must be configured more aggressively, allowing the buy rates to rise higher than the market maker rate in order for the pump to remain operational. Finally, the large demonstration study, performed as part of WP5 activities, confirms the applied functionality and the value of the HYPERGRYD-supported ICT tool development (heat pump and district heating digital twins development and integration in the Grid Singularity Exchange). The study shows how the GSY Simulation Tool [3] and the backend code of the Grid Singularity Exchange [2], enhanced by research and development conducted in this project, can be deployed to analyse the potential of sector coupling and flexibility assets (notably heat pumps), and how renewable energy sources, combined with district heating systems, can be further harnessed by enabling peer-to-peer trading for local energy communities and implementing dynamic grid tariffs. The final version of the report (D3.6: Description and report of energy exchange and smart trading tool for LEC including simulation studies for living labs as active LECs. Final version) will include the feedback from SONNE users and other stakeholders, including regulators, provided as part of final activities of T5.5, as well as in related dissemination and exploitation activities in WP6 - Communication and Dissemination. D3.5 Description and report of energy exchange and smart trading tool including simulation studies for living labs as active LECs. Preliminary version 99 References [1] HYPERGRID, Hybrid Coupled Networks for Thermal-Electric Integrated Smart Energy Districts, https://hypergryd.eu/, (accessed 16.05.2023) [2] Grid Singularity Exchange, with backend source code at https://github.com/gridsingularity/gsy-e and documentation including technical approach and licensing described in GSY Wiki, https://gridsingularity.github.io/gsy-e/licensing/ (accessed 23.11.2023) [3] Singularity Map, https://gridsingularity.com/singularity-map/, (accessed 11.11.2023) [4] Grid Singularity Wiki, Heat Pumps and District Heating, https://gridsingularity.github.io/gsy-e/heat-pumps-general/, (accessed 23.11.2023) [5] Grid Singularity Wiki, Results Dashboard (Key Performance Indicators), https://gridsingularity.github.io/gsy-e/results-dashboard/, (accessed 11.05.2023) [6] Coefficient of Performance, https://www.sciencedirect.com/topics/engineering/coefficient-of-performance, (accessed 11.05.2023) [7] FEDECOM Project, https://fedecom-project.eu/, (accessed 16.05.2023) [8] Grid Singularity Wiki, Heat Pump Trading Strategy and Asset Configuration Options, https://gridsingularity.github.io/gsy-e/heat-pump/, (accessed 11.11.2023) [9] Grid Singularity Wiki, Market Types Overview, https://gridsingularity.github.io/gsy-e/market-types/, (accessed 11.11.2023) [10] Grid Singularity Wiki, Virtual Heat Pump, https://gridsingularity.github.io/gsy-e/virtual-heat-pump/, (accessed 11.11.2023) [11] Grid Singularity Wiki, Trading Strategies, https://gridsingularity.github.io/gsy-e/default-trading-strategy/, (accessed 11.05.2023) [12] Bordignon, S., Spitler, J.D., Zarrella, A. Simplified Water-Source Heat Pump Models for Predicting Heat Extraction and Rejection, 2024, DOI: 10.1016/j.renene.2023.119701, https://www.sciencedirect.com/science/article/pii/S0960148123016166 (accessed 09.07.2023) [13] Grid Singularity Wiki, Data Requirements and File Upload Format, https://gridsingularity.github.io/gsy-e/data-requirements/, (accessed 11.05.2023) [14] European Union, General Data Protection Regulation, https://eur-lex.europa.eu/eli/reg/2016/679/oj, (accessed 11.05.2023) [15] The Copernicus Climate Change Service, https://climate.copernicus.eu/, (accessed 11.08.2023) and Hersbach, H., Bell, B., Berrisford, P., Biavati, G., Horányi, A., Muñoz Sabater, J., Nicolas, J., Peubey, C., Radu, R., Rozum, I., Schepers, D., Simmons, A., Soci, C., Dee, D., Thépaut, J-N. (2023): ERA5 hourly data on single levels from 1940 to present. Copernicus Climate Change Service (C3S) Climate Data Store (CDS), DOI: 10.24381/cds.adbb2d47, https://cds.climate.copernicus.eu/cdsapp#!/dataset/reanalysis-era5-single-levels (accessed 09.07.2023) [16] Grid Singularity Wiki, Solar Panels (PVs), https://gridsingularity.github.io/gsy-e/solar-panels/, (accessed 11.05.2023) [17] EnergyDataMap, https://mapped.energy/, (accessed 11.07.2023) [18] Rebase Energy, The Python-first energy forecasting platform, https://www.rebase.energy/ (accessed 11.07.2023) [19] Grid Singularity Wiki, Asset Configuration Overview, https://gridsingularity.github.io/gsy-e/configuration/, (accessed 11.11.2023) [20] Grid Singularity Wiki, Two-Sided Pay-as-Bid Market, https://gridsingularity.github.io/gsy-e/two-sided-pay-as-bid/, (accessed 11.05.2023) [21] SonnenShop, BYD Battery-Box H 11.5, https://www.sonnenshop.de/speichersysteme/byd/byd-battery-box-h-11.5, (accessed 11.05.2023) [22] EUROSTAT, Heating and cooling from renewables gradually increasing, 3 February 2023, https://ec.europa.eu/eurostat/web/products-eurostat-news/w/DDN-20230203-1, (accessed 16.05.2023) [23] Lyons, L., Georgakaki, A., Kuokkanen, A., Letout, S., Mountraki, A., Ince, E., Shtjefni, D., Joanny, G., Eulaerts, O.D. and M. Grabowska, Clean Energy Technology Observatory: Heat Pumps in the European Union – 2022 Status Report on Technology Development, Trends, Value Chains and Markets, Publications Office of the D3.5 Description and report of energy exchange and smart trading tool including simulation studies for living labs as active LECs. Preliminary version 100 European Union, Luxembourg, 2022, https://publications.jrc.ec.europa.eu/repository/handle/JRC130874 (accessed 09.05.2023) [24] European Union, Horizon Europe, 2021-2027 Research and Innovation Programme, https://research-and-innovation.ec.europa.eu/funding/funding-opportunities/funding-programmes-and-opencalls/horizon-europe_en (accessed 09.05.2023) [25] European Union, Recovery and Resilience Programme, https://commission.europa.eu/business-economy-euro/economic-recovery/recovery-and-resilience-facility_en, (accessed 09.05.2023) [25] Grid Singularity Wiki, Setup Configuration, https://gridsingularity.github.io/gsy-e/setup-configuration/ (accessed 11.05.2023) D3.5 Description and report of energy exchange and smart trading tool including simulation studies for living labs as active LECs. Preliminary version 101