Dissemination
Publications
- Araveti, Sandeep, Cristian Aguayo Quintana, Evita Kairisa, Anna Mutule, Juan Pablo Sepulveda Adriazola, Conor Sweeney, and Paula Carroll. 2022. “Wind Energy Assessment for Renewable Energy Communities” Wind 2, no. 2: 325-347. . https://doi.org/10.3390/wind2020018
- T. Korõtko, I. Drovtar, A. Mutule, E. Kairisa and A. Rosin, “Load Flow Modelling in Local Energy Community Electric Power Systems,” 2022 IEEE 7th International Energy Conference (ENERGYCON), Riga, Latvia, 2022, pp. 1-7 https://ieeexplore.ieee.org/abstract/document/9830203
- I. Drovtar, T. Korõtko, A. Mutule, E. Kairisa and A. Rosin, “Determining optimisation Framework for Local Energy Communities,” 2022 IEEE 7th International Energy Conference (ENERGYCON), Riga, Latvia, 2022, pp. 1-7 https://ieeexplore.ieee.org/abstract/document/9830444
- E. Kairisa, A. Mutule, I. Drovtar, T. Korõtko, O. Borscevskis, H.Wilkening, C.Troyer // Scenario-based Modelling of Residential Sector Consumption: A Case Study in Latvia // pieņemts un tiks publicēts LATVIAN JOURNAL OF PHYSICS AND TECHNICAL SCIENCES 2022, N 3. https://doi.org/10.2478/lpts-2022-0014
- Korõtko, T.; Plaum, F.; Häring, T.; Mutule, A.; Lazdins, R.; Borščevskis, O.; Rosin, A.; Carroll, P. Assessment of Power System Asset Dispatch under Different Local Energy Community Business Models. Energies 2023, 16, 3476 https://doi.org/10.3390/en16083476
- Zahraoui, Y.; Korõtko, T.; Rosin, A.; Ahmadiahangar, R. Stochastic Energy Management for Battery Storage System Based Microgrid Considering Different Forecasting Models. Accepted for publication in CPE-POWERENG 2023 https://doi.org/10.1109/CPE-POWERENG58103.2023.10227451
- Agabus, H.; Korõtko, T.; Kull, K.; Rosin, A.; Palu, I. Potential Assessment of Closed Distribution System Uptake in Estonia. Accepted for publication in CPE-POWERENG 2023 https://doi.org/10.1109/CPE-POWERENG58103.2023.10227469
- Korõtko, T.; Zahraoui, Y.; Rosin, A.; Agabus, H. Digital Twins for Designing Energy Management Systems for Microgrids: Implementation Example Based on TalTech Campulse Project. Accepted for publication in CPE-POWERENG 2023 https://doi.org/10.1109/CPE-POWERENG58103.2023.10227475
Conference presentations and other events
-
IFORS 2023, the 23rd Conference of the International Federation of Operational Research Societies, July 10-14, 2023, Santiago, Chile.
-
ORBEL 2023, May
-
ROADEF 2023, February
-
Cristian Aguayo, Paula Carroll, Bernard Fortz, Evita Kairiša, Juan Pablo Sepulveda Adriazola, OR in Energy - Supporting Energy Communities, EURO 2022, the 32nd EURO Conference of the Association of European Operational Research Societeis, July 3-6, 2022, Espoo, Finland.
-
PMGO 2022, November
-
Paula Carroll, Cristian Aguayo, Sandeep Araveti, Luce Brotcorne, Imre Drovtar, Bernard Fortz, Evita Kairiša, Tarmo Korõtko, Anna Mutule, and Conor Sweeney, Supporting Energy Communities - Operational Research and Energy Analytics, INOC 2022, the International Network Optimization Conference, June 7-10, 2022, Aachen, Germany
-
Sandeep Araveti, Conor Sweeney, and Paula Carroll, Identifying the best wind and weather datasets for local energy models in Ireland, WEI 2022, Wind Energy Ireland Annual Conference, May 12th 2022.
Workshops
Kickoff Workshop, University College Dublin, November 26-27, 2021
Université libre de Bruxelles, September 1-2, 2022
INRIA Lille, January 26-27, 2023
Riga Technical University, October 9-10, 2023
Reports
1.1 Data management plan
1.2 Visualization platform V1
1.3 Platform + Data archive
2.1 Climate RES scenarios
2.2 LEC/SM demand scenarios
2.3 LEC design options.
3.1 Model1-Bilevel DSM – No Sharing
3.2 Model1 Algorithms
3.3 Model2-Bilevel DSM – LEC sharing
3.4 Model2 Algorithms
4.1 Deterministic & decomposition techniques
The Unit Commitment (UC) problem is one of the typical approaches to address the energy dispatch problem from the point of view of a DSO in order to satisfy the energy demand in such a way that operational costs are minimised. This problem involves decisions about a set of power generating units in terms of their operation in different periods. The decisions involved in this problem are divided into two: discrete decisions, related to the switching on and off of the generating units in different time periods; and continuous decisions, related to how much energy the generating units produce in each period.
The mentioned structure suggests a possible solution strategy called Benders’ decomposition. The idea behind this strategy is to solve a problem that considers only discrete decisions, their respective constraints and their respective costs, and then solve a second linear problem that considers only continuous decisions using as information the discrete decisions taken previously, including constraints for these decisions and including costs associated with them. The second problem has two possible outcomes of interest:
- The problem is infeasible: using duality properties in linear programming, it is possible to generate feasibility constraints (called feasibility cuts) which must be added to the discrete decision problem and obtain a different solution.
- The problem has a finite optimal solution: in this case, this finite solution corresponds to a lower bound of the costs of continuous decisions, so constraints (optimality cuts) must be generated in the discrete decision problem to ensure a cost equal to or greater than that obtained by the problem with continuous decisions.
4.2 Prototype Algorithms
For benchmark instances of the UC problem, the solving time using Bender’s decomposition was compared to the solving time solving the UC problem with a commercial solver with access to academic licenses. The result was that the commercial solver solves these instances much faster than the Benders’ decomposition.
4.3 Robust model with scenario generation
The main constraint of the UC problem is the satisfaction of energy demand for each period. In theory, energy demand is uncertain, and on the other hand, if renewable energy sources are to be included in the problem, their availability is also uncertain. That is why a strategy to deal with uncertainty must be adopted. Robust optimization is a subtopic of optimization that deals with uncertainty. To adequately address the UC problem using a robust optimization approach, it is necessary to have historical data to be able to approximate the necessary parameters and generate likely scenarios for both energy demand and renewable energy availability.
4.4 Algorithm implementation
To solve the robust version of the UC problem, two solution strategies were implemented:
- Consider the problem as a two-stage problem: very similar to Benders' decomposition, discrete decisions are proposed to be first-stage decisions, and then continuous decisions are considered as second-stage decisions. This requires a set of energy demand and renewable energy availability scenarios, so that the second-stage decisions are feasible for all scenarios in the set of scenarios.
- Reformulate the problem in such a way as to obtain a fully adaptive decision policy: By reformulating the continuous decisions as (linear) functions of the different energy demand and renewable energy availability scenarios, solutions are obtained that allow adjusting the energy production depending on the variations resulting from the scenarios.
Preliminary results show that the problem statement as a two-stage problem can be quickly solved using a column and constraint generation method. However, the two-stage approach may be infeasible in the context of operational decision making, unlike the adaptive approach.
5.1 Reference system architecture
The report delineates a reference system architecture tailored for Local Energy Communities (LECs), comprising literature reviews on legal structures, performance metrics for LEC modeling, and key steps for digital twin development. It delves into legal forms of LECs, analyzing their purpose, membership, profit application, and includes case study examples. Additionally, it outlines a general optimization framework and a universal prosumer model.
Exploring diverse legal forms of LECs in Section 2, it scrutinizes their purpose, membership structures, and profit applications, illustrated through case studies. Section 3 details the general framework for LEC mathematical optimization models and the universal prosumer model, aiming for a versatile model adaptable to various community sizes. It elucidates the local energy community concept, its infrastructure model, scenario development methodology, and objectives like cost reduction, energy production, and carbon footprint reduction.
Three primary objectives for LECs are maximizing renewable energy production, minimizing impact on grid connection capacity, and boosting self-consumption. Constraints include grid connection availability, shiftable load, and energy storage. Section 3.3 ranks scenarios based on effectiveness indicators, showcasing the correlation between CO2 reduction and locally consumed renewable energy. It highlights the positive impact of demand-side response and energy storage on self-consumption and system optimization.
Section 3.4 demonstrates the feasibility of a universal prosumer modeling object for simulating steady-state load-flow in LEC electric power systems. It employs custom control blocks and interfaces to streamline power flow modeling, reducing resource requirements. Section 3.5 creates an electric power system model for steady-state active power distribution within an LEC, utilizing MATLAB Simulink for a configurable load-flow model with interfaces to external software.
Sections 4 and 5 expound on the V-model methodology and digital twin development, complementing each other in defining reference architecture. Overall, the report emphasizes integrating technologies for energy efficiency in LECs, with the universal prosumer model facilitating versatile simulations for holistic operational analysis.
