Multi swarm optimization based adaptive fuzzy multi agent system for microgrid multi-objective energy management

Micro grids (MG) are seen as the future power system providing clear economic and environmental benefits. Most MG energy management solutions rely on centralized controller which is not suitable to guarantee the flexibility and the adaptability that modern electricity market needs. In other hand, multi objective optimization for fully decentralized system in MG environment is not realizable without certain level of coordination between agents. In this paper, we present an adaptive multi-agents system (AMAS) for MG power management based on enhanced fuzzy decision using multi swarm optimization (MS-PSO) algorithm. In the proposed architecture each agent presents a different MG unit. Fuzzy logic is used by each agent to estimate the amount of energy to be generated in order to cover the uncertainty and imprecision related to renewable energy sources and the MG constraints. For the MAS coordination, a MS-PSO algorithm is used by a coordinator agent to find the best compromised solution to satisfy economical/environmental objective based on agent proposals in order to improve them. Simulation results show the importance of the chosen optimization algorithm for the AMAS with MS-PSO algorithm which is compared to the basic particle swarm optimization for the same encapsulated knowledge.