A new self adaptive particle swarm optimization technique for optimal design of a hybrid power system

Present paper discusses about a reliable and cost effective optimal design for hybrid energy systems using meta-heuristic techniques. Design, planning and control of hybrid systems require complex optimization of the same which may not be solved easily with the conventional optimization methods. Several meta-heuristic optimization techniques such as particle swarm optimization techniques (PSO), differential evolution (DE), genetic algorithm (GA) etc have been successfully applied to solve these problems. One of the major difficulties in these methods is the premature convergence. In the present paper a new improved optimization technique based on PSO has been proposed to avoid premature convergence while optimizing the overall cost of energy of a hybrid power system of wind turbine, photovoltaic generator, diesel generator and battery bank. The results obtained by this improved method are compared with an iterative method. It is found that the new improved method can produce superior results.