一种新的群集智能优化及其改进研究

Novel swarm intelligence algorithm and its improvement

研究一种新的群集智能优化算法—自由搜索(FS)算法。提出了该算法的改进策略,实时调整个体的邻域搜索半径和精英保留。用典型测试函数对FS的改进算法和微粒群算法(PSO)进行对比实验,实验结果验证了算法的正确性和高效性。该算法不仅在收敛精度、收敛速度方面较PSO算法有明显的提高,而且全局搜索能力更强。

A novel simulated evolutionary algorithm, free search, is studied. An improved algorithm of FS including changing search neighbor areas dynamically and elitist strategy are presented. Four benchmark functions are tested and the experimental results show that the new algorithm not only significantly speed up the convergence, but also effectively solve the premature convergence problem. Comparing with particle swarm optimization, the performance of the proposed algorithm is greatly improved consequently.