A Novel Reinforcement Learning-Based Multi-Objective Energy Management System for Multi-Energy Microgrids Integrating Electrical, Hydrogen, and Thermal Elements

Abstract

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, and hydrogen systems within a multi-energy microgrid (MEMG). The EMS employs a reinforcement learning (RL) algorithm (RL-MOF-EMS) to achieve the energy balance, ensuring appropriate distribution of thermal and electrical power. Three distinct single-objective functions are considered: 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 swarm optimization (PSO). A multi-objective problem combining these three objectives is formulated and the RL-MOF-EMS is implemented to determine the optimal operation of the MEMG. The performance is rigorously tested in Simulink under diverse weather conditions and fluctuating2 thermal and electrical demand profiles. The results are compared against an EMS based on a nonlinear MATLAB optimizer, demonstrating the effectiveness of the RL-MOF-EMS in coordinating power flows from energy sources and storages. This coordination minimizes costs and component degradation, while meeting energy demands in all scenarios. The findings indicate that the MEMG operates with a high degree of self-sufficiency, reducing reliance on the grid