Particle swarm optimization with opposition-based learning and near neighbor interactions


Jin Wang - School of Information, Linyi University, Linyi 276000, China.


Particle swarm optimization (PSO) is recently proposed as population-based stochastic algorithm, which has shown excellent abilities in many optimization problems. In this paper, a hybrid PSO variant is presented to enhance its performance. The new algorithm is called OFDR-PSO which employs opposition-based learning (OBL) and fitness-distance-ratio (FDR). In order to verify the performance of OFDR-PSO, we test in on a set of well-known benchmark problems. Simulation results demonstrate that our proposed approach is effective and outperforms other four compared algorithms.