Adaptive PSO Algorithm With Non-Linearly Decreasing Inertia Weight

This paper proposes a modified particle swarm optimization method with non linearly decreasing inertia weight (MPSO-NDIW) and time varying acceleration coefficients. In this MPSO-NDIW method, proper control of the global exploration and local exploitation is done in finding the optimum solution efficiently. In the early stage, full range of search space is allowed for search by the PSO algorithm and in the later stages of search, fine tuning of solution is done so that the algorithm converges to the global optima efficiently .In velocity vector equation inertia weight and acceleration coefficients are non-linearly varied with iterations. To compare the performance of the MPSO-NDIW method with the other improved PSO, four well known benchmark test functions are used. The results reveal that MPSO-NDIW is a efficient technique and has better performance.