RT journal article
T1 A Novel Reinforcement Learning-Based Multi-Objective Energy Management System for Multi-Energy Microgrids Integrating Electrical, Hydrogen, and Thermal Elements
A1 Hosseini, Ehsan
A1 García Triviño, Pablo
A1 Horrillo Quintero, Pablo
A1 Carrasco González, David
A1 García Vázquez, Carlos Andrés
A1 Sarrias Mena, Raúl
A1 Sánchez Sainz, Higinio
A1 Fernández Ramírez, Luis Miguel
A2 Ingeniería Eléctrica
A2 Ingeniería en AutomáticaElectrónica, Arquitectura y Redes de Computadores
K1 Multi-energy microgrids
K1 reinforcement learning
K1 energy management system
K1 electricity
K1 hydrogen
K1 thermal
AB This paper presents an advanced multi-objective function based-energy management system(MOF-EMS) designed to optimize the economic dispatch and lifespan of electrical, thermal, andhydrogen systems within a multi-energy microgrid (MEMG). The EMS employs a reinforcementlearning (RL) algorithm (RL-MOF-EMS) to achieve the energy balance, ensuring appropriatedistribution of thermal and electrical power. Three distinct single-objective functions areconsidered: 1) minimizing operational cost, 2) reducing the life degradation of the storage devices,and 3) minimizing heating cost. Each objective is evaluated under three different EMSs based on:1) priority-based regulator (PBR), 2) proportional regulator (PR), and 3) particle swarmoptimization (PSO). A multi-objective problem combining these three objectives is formulated andthe RL-MOF-EMS is implemented to determine the optimal operation of the MEMG. Theperformance is rigorously tested in Simulink under diverse weather conditions and fluctuating2thermal and electrical demand profiles. The results are compared against an EMS based on anonlinear MATLAB optimizer, demonstrating the effectiveness of the RL-MOF-EMS incoordinating power flows from energy sources and storages. This coordination minimizes costsand component degradation, while meeting energy demands in all scenarios. The findings indicatethat the MEMG operates with a high degree of self-sufficiency, reducing reliance on the grid
PB Elsevier
YR 2025
FD 2025-01-31
LK http://hdl.handle.net/10498/35296
UL http://hdl.handle.net/10498/35296
LA eng
NO This work was partially supported by Ministerio de Ciencia e Innovación, Agencia Estatal deInvestigación, and Unión Europea “NextGenerationEU/PRTR” (Grant TED2021-129631B-C32supported by MCIN/AEI/10.13039/501100011033 and NextGenerationEU/PRTR)
DS Repositorio Institucional de la Universidad de Cádiz
RD 09-may-2026