Comprehensive learning particle swarm optimization with Tabu operator based on ripple neighborhood for global optimization

Abstract

For the weak convergence at the latter stage of the comprehensive learning particle swarm optimizer (CLPSO), we put forward a new CLPSO based on Tabu search to enhance the performance. Inspired by the phenomenon of water waves, a Ripple Neighborhood (RP) structure based on the Gaussian distribution is proposed to construct a new adaptive neighborhood structure to guide the selection of candidate solutions in Tabu search, which solves the problem of low convergence and improves the quality of the solution in CLPSO. Experimental results on the standard 26 test functions show that the proposed algorithm achieves a better performance compared with CLPSO.