一种改进的混合蛙跳算法及其收敛性分析  

Improved shuffled frog-leaping algorithm and its convergent analysis

作　　者：贺毅朝[1] 曲文龙[1] 许冀伟[1] 

机构地区：[1]石家庄经济学院信息工程系,石家庄050031

出　　处：《计算机工程与应用》2011年第22期37-40,共4页Computer Engineering and Applications

基　　金：国家自然科学基金No.10971052;河北省教育厅青年基金项目(No.2010260)~~

摘　　要：为了提高混合蛙跳算法(SFLA)求解函数优化问题的能力,借鉴PSO与DE的进化算子提出了一种改进的混合蛙跳算法(ESFLA),分析了ESFLA的时间复杂性,并基于有限Markov链证明了ESFLA的全局收敛性。对ESFLA、SFLA与ISFLA2的仿真计算结果表明,ESFLA比SFLA和ISFLA2更适用于求解复杂的函数优化问题。To improve the ability of shuffled frog-leaping algorithm（SFLA）for solving function optimization problems,an improved efficiently shuffled frog-leaping algorithm（ESFLA） is proposed,which adopts the evolutionary methods of particle swarm optimization and differential evolution,and its time complexity is analysed.The global convergence of ESFLA is proved by using limit Markov chain.In order to test and verify the ability of ESFLA for solving the function optimization problems,the performance of ESFLA is compared with that of SFLA and ISFLA2.The experimental results indicate that the performance of ESFLA is superior to SFLA and ISFLA2,and it is more suitable for solving complex function optimization problems.

关 键 词：混合蛙跳算法 粒子群优化 差分演化 全局收敛性 函数优化 

分 类 号：TP301.6[自动化与计算机技术—计算机系统结构]