混沌增强加速粒子群优化算法  

Chaos-enhanced accelerated particle swarm optimization algorithm

作　　者：赵丽萍[1] 舒期梁[1] 武燕[1] 李孟山[1] 

机构地区：[1]景德镇陶瓷学院信息工程学院,江西景德镇333001

出　　处：《计算机应用研究》2014年第8期2307-2310,共4页Application Research of Computers

基　　金：国家自然科学基金资助项目(61202313)

摘　　要：针对最近提出的具有极强全局搜索能力的加速粒子群算法,为改善早熟收敛问题并提高收敛精度,提出一种融合混沌理论的混沌增强加速粒子群算法。该算法引入混沌序列来调节全局学习因子,使算法进一步增加全局搜索能力。算法性能通过测试四个典型多目标优化函数验证,并与经典的非劣分类多目标遗传算法、多目标粒子群优化算法和加速粒子群算法相比较,结果表明混沌增强加速粒子群算法具有较快的收敛速度和较强的跳出局部最优能力,性能优越,可供优化求解等许多领域借鉴。According to the recently proposed accelerated particle swarm optimization（ APSO） algorithm which showed extra advantages in convergence for global search,to improve its premature and convergence accuracy,this paper developed a novel chaos-enhanced APSO based on chaos theory,hereafter called CAPSO. In CAPSO,it proposed chaotic sequence for tuning the global learning factor in order to further enhance its global search ability. The performance of the proposed algorithm was carried out on four classical multi-objective optimization functions by comparing with the non-dominated sorting genetic algorithm,multi-objective PSO and APSO. The results verify that the CAPSO algorithm shows superior performance with rapid convergence and stronger ability to jump out of local optimum,and can provide reference for lots of optimization and computation field.

关 键 词：粒子群优化 混沌增强 加速算法 多目标优化 

分 类 号：TP301.6[自动化与计算机技术—计算机系统结构]