一种混合粒子群优化算法的研究  

Research on hybrid particle swarm optimization algorithm

作　　者：张安玲[1] 王中[2] 

机构地区：[1]长治学院数学系,山西长治046011 [2]长治学院计算机系,山西长治046011

出　　处：《计算机工程与应用》2011年第31期27-29,37,共4页Computer Engineering and Applications

基　　金：长治学院科研项目(No.2009010)

摘　　要：结合粒子群优化算法和拟牛顿法的优点,提出了一种混合粒子群优化算法。该算法首先运行粒子群优化算法,到进化到一定程度时,把当代的最好点作为拟牛顿法的初始点,再利用拟牛顿法,对其进行二次优化。算法充分发挥了粒子群优化算法的全局搜索性和拟牛顿法的局部精细搜索性,同时也克服了粒子群算法后期搜索效率低和拟牛顿法对初始点敏感的缺陷。数值实验结果表明,该算法具有很高的收敛速度和求解精度。A hybrid particle swarm optimization algorithm combing advantages of Particle Swarm Optimization（PSO） algorithm with quasi-Newton method is proposed.The algorithm runs the PSO firstly.The best point of contemporary is used as the initial point of quasi-Newton method when PSO evolutes to a certain extent.Then the algorithm is further optimized using quasi-Newton method.The hybrid algorithm has displayed sufficiently the characteristics of PSO＇s group search and quasi-Newton method＇s local strong search.At the same time,it overcomes the disadvantages of high sensitivity to initial point of quasi-Newton method and PSO reducing the search efficiency in later period.Numerical results show that the algorithm has a high convergence speed and solution precision.

关 键 词：粒子群优化算法 拟牛顿法 混合算法 收敛性 

分 类 号：O224[理学—运筹学与控制论] TP18[理学—数学]