Micro grid renewables dynamic and static performance optimization using genetic algorithm

Abstract

This paper presents Microgrid (MG) optimization using Genetic Algorithm. The MG model is based on renewable energy sources (wind turbine) and gas generator. The algorithm objective is determine the optimal size of combined wind and gas generator to satisfy a given Key Performance Indices (KPIs). The selected KPIs describe both dynamic and static performance of MG. The KPIs describing the dynamic performance includes Total Harmonic Distortion (THD) and power factor (PF) in presence of disturbance and load variation. The static KPIs includes power shortage, initial cost, running cost and CO2 emission. The two KPIs groups (dynamic and static) have different time frame. The dynamic KPIs are examined by applying load disturbance to MG and observe its effect over few seconds (according to MG average time constant). The static KPIs are examined by applying load and power generation profiles during one full year period. Hence, it is not feasible to combine both static and dynamic simulation using one model. Accordingly, to allow one optimization process based on static and dynamic KPIs, two simulation models have been created with two separate simulation environments. The static simulation uses simplified efficiency model of the power components presented in MG and the system is subjected to load and wind profiles to evaluate the static KPIs. The dynamic simulation uses detailed dynamic model with load disturbance. The optimization process utilizes a single fitness function which combines the dynamic and static PKIs with weighting factors. Results of optimization are presented and the KPIs of the optimized MG is provided.