Opposition based modified particle swarm optimization algorithm
Om Prakash Verma, Shivang Gupta, Shubham Goswami, Swapan Jain · 2017
A modified version of Particle Swarm Optimization algorithm(PSO) originally proposed by Eberhart and Kennedy is presented in this paper. The proposed algorithm is different from the original PSO algorithm in two respects: Firstly, opposition based technique is employed in which the initial particles are generated using opposition based learning to increase convergence towards the optimal solution. Secondly, a novel dimension based learning approach is used in which each dimension is considered separately for finding the global optimal solution. Applying these methods leads to an optimum solution for an objective function having a large number of dimensions. The algorithm has been tested for several complex multidimensional mathematical functions for experimental verification.