Particle swarm optimization with selective multiple inertia weights

Particle swarm optimization is widely used in past decades as optimization method for unimodal, multimodal, separable and non-separable optimization problems. A popular variant of PSO is PSO-W (Inertia Weight PSO). Attempts has made to modify the PSO with Selective Multiple Inertia Weights (SMIWPSO) to enhance the searching capability of PSO. The present paper implemented the SMIWPSO with four best chosen Inertia Weight techniques i.e. Linear Decreasing Inertia Weight, Chaotic Inertia Weight, Random Inertia Weight and Constant Inertia Weight. Selection of considered Inertia Weight depends upon the agreement of controlling parameter P. SMIWPSO performance is examine against PSO with respect to 25 standard optimization problem. Experimental results show SMIWPSO have significant improvement in relation to efficiency, reliability and robustness.