Multi Agent System Based Decentralized Approach for Smart Energy Management in a Hybrid Microgrid /

Abstract

Renewable-based microgrid systems can play a major role in the energy sector by addressing increasing power demand and climate change challenge issues. The proposed study considers a hybrid microgrid using Renewable Energy Sources (RES), diesel generator, and energy storage system. Three cases of microgrid operation are considered: islanded mode, transition to grid mode and grid-oriented mode. The hybrid microgrid operates in each mode considering the weather pattern, energy costs, nature of the connected sources and load demand. A robust and automated Multi Agent System (MAS) for controlling the microgrid operation is used. In the islanded mode, four climatic scenarios are considered: best, extreme sunny, extreme cloudy, and worst climate. The transition to grid mode occurs when the climate is sunny and cloudy and the battery is not enough to support the load demand. In grid-oriented mode, two cases include grid-tied and grid-feed mode. In grid feed mode, the microgrid has surplus energy and sells it to the microgrid. In grid-tied mode, the RES and the battery energy system are insufficient to support the load demand. The microgrid becomes fully dependent on the utility grid by buying power from it and supporting the load demand through the utility grid. Furthermore, for the optimization of the microgrid power and energy cost a metaheuristic Particle Swarm Optimization (PSO) technique is implemented. The study results indicate the cost-effectiveness of RES utilized by the microgrid as compared to the utilization of fossil fuel sources and utility grid.