A Voronoi Decomposing Cooperative Learning Particle Swarm Optimization with Random Walk Strategy

Cooperative Co-evolutionary Particle Swarm Optimization (CCPSO) is based on a cooperative co-evolutionary framework and has been proposed for solving large scale nonseparable problems. This algorithm uses random grouping strategy for decomposing variables. In this paper we propose a Voronoi-bacesd decomposition for Cooperative Learning Particle Swarm Optimization (VCLPSO). Instead of random grouping, the proposed approach uses Voronoi decomposing strategy and then in order to increase the explorability of the algorithm in a multi modal search space we use Cauchy and Gaussian distributionsand random walk. These two techniques yield to faster and more mature convergence of the algorithm. The results are evaluated using several CEC2010 benchmarks, and also are compared with results of several state-of-the-art approaches such as cooperation co-evolution particle swarm optimization (CCPSO)

A Voronoi Decomposing Cooperative Learning Particle Swarm Optimization with Random Walk Strategy Keywords:

Cooperative , Adaptive Learning , Voronoi Decomposing , large scale optimization , Random Walk