Abstract:
Rapid decline of fossil fuel reserves and rise of average global temperature has compelled energy scientists to look for non-conventional energy sources, preferably environment friendly and renewable in nature. Among the renewable sources, wind and photovoltaic based energy conversion processes are capturing recent interests. As the input to these two kinds of energy conversion processes are highly unpredictable, incorporation of energy storage device becomes imperative for uninterruptible power supply. However, considering hybrid renewable power generation for fulfilling load demand, arbitrary mixing among participating generating units could result in non-profitable outcome for power supplying entities. Hence, in this work, an optimal sizing of a Wind-Photovoltaic-Battery system has been suggested using a hybrid single objective optimization (SOO) method integrating a genetic algorithm (GA) and grey wolf optimizer (GWO) in phase one and in phase two a multi-objective optimization (MOO) method integrating a non-dominant sorting Genetic Algorithm (NSGA) II and the Grey wolf optimizer (GWO) is proposed. In the SOO phase the population undergoes cross-over and mutation and then the population is updated according to GWO. In the MOO phase the population of each generation of NSGA II is passed through the GWO before they are allowed to crossover and mutate in order to increase the probability of avoiding local minima. A comparative analysis of the performance of the applied hybrid algorithm with NSGA II and multi-objective Particle Swarm Optimization (MOPSO) has been carried out in phase two and the hybrid SOO algorithm in phase one is compared with GA. The analysis shows that the applied hybrid algorithms show better performance compared to the other existing algorithms in terms of convergence speed, obtaining global minima, lower mean (for minimization objective) and a higher standard deviation.
