PERFORMANCE ANALYSIS OF PARTICLE SWARM OPTIMIZATION APPROACH FOR OPTIMIZING ELECTRICITY COST FROM A HYBRID SOLAR, WIND AND HYDROPOWER PLANT

This paper deals with the cost optimization of a hybrid solar, wind and hydropower plant using a Particle Swarm Optimization (PSO) approach. PSO is a technique that belongs to Swarm intelligence, an artificial intelligence (AI) technique, known as a Meta-heuristic optimization solver, mostly used in Biology. With the consideration of solar, wind and hydro hybrid system which has become extremely relevant for developing countries, and also the existence of a wide list of constraints, the adoption of PSO technique cannot be avoided. On the other hand, a linear optimization approach was used with Matlab software to solve the same problem. Both techniques were applied to secondary data collected from RetScreen Plus software for the location Accra and results were extracted in terms of distribution of supply by individual sources and cost of hybrid system electricity. Results show in general, an improvement of hybrid system cost of electricity. A histogram was used to show the distribution of supply for the particular load and the equivalent cost of hybrid system that corresponds to it. A khi-sqaure test ws run to compare the two series of data obtained from the two approaches adopted. The Khi-square test show high similarity confirming the reliability of the PSO approach which on the other hand presents the advantage of scalability over a wider range of sources with multiple constraints.