An Enhanced Opposition-Based Particle Swarm Optimization
Jun Tang, Xiaojuan Zhao · 2009
Particle swarm optimization (PSO) has shown its fast search speed in many optimization and search problems. However, PSO easily fall into local optima on some multimodal and complicated problems. In this paper, an enhanced opposition-based PSO, called EOPSO, is proposed by combing an enhanced opposition-based learning and the standard PSO. The enhanced opposition provides solutions more closely to the global optimum than the traditional opposite solutions. Experimental studies on 4 unimodal functions and 4 multimodal functions show that the EOPSO does not only surpass the standard PSO and opposition-based PSO on all test functions, but also shows faster convergence rate.