Optimization of Droop Control Based Islanded Microgrid System under Load Variation

Abstract

In an islanded microgrid, the system operating frequency and voltage is governed by the socalled droop based relationships due to the absence of conventional slack bus which necessitates incorporation of various controllers. The parameters of the controllers play a vital role ensuring stable and reliable supply of power to the connected consumers. In this work, a two-stage optimal tuning procedure is presented for controlling a two inverter based microgrid system under isolated mode of operation. At rst, the key control parameters in uencing the transient response of the system are identi ed. Next, an eigenvalue based optimization process has been adopted in which three population based search algorithms: Genetic Algorithm (GA), Particle Swarm Optimization (PSO) algorithm and Grey Wolf Optimization (GWO) algorithm are used in order to nd out the initial range of controller parameters in an unbounded search space. In the second stage of optimization, Exhaustive Search (ES) and Interior Point (IP) based search algorithms are adopted to obtain the optimized value within the bounds provided by the population based search algorithms. At the end, time domain and comparative analyses among these combinational algorithms are presented to determine the most e ective algorithm for the studied islanded microgrid system. The simulation results show that the performance of GA is inferior compared to other two algorithms: PSO and GWO. GWO provides better optimization in comparatively lesser amount of time than PSO in the rst stage and in the second stage, IP yields better optimization than ES. Consequently, the two stage optimization process suggests that GWOIP could be chosen for the design problem.