摘要
为了提高微粒群算法优化高维目标的性能,采用了个体惯性权重自适应调整的微粒群算法,其中每个微粒拥有属于个体的惯性权重。通过对每个微粒的适应值进行评价对惯性权重动态和自适应,以加快其收敛速度并逃离局部最优。为了增强搜索性能,基于高斯变异和随机变异的变异算子被引入。该方法以及其他3种不同微粒群优化算法对4个经典函数在100、200和400维数下进行仿真的结果比较证明此算法在解决高维数目标时具有良好性能。
To enhance the performance of the particle swarm optimization （PSO）, the self-adaptive individual inertia weight adjustment particle swarm optimization was employed. In the approach each particle has an individual inertia weight in the approach. The individual inertia weight will be adjusted dynamically and self-adaptively by evaluating the fitness value of the passed evolutions to speed up convergence and escape local optima. Moreover, a mutation operator was employed based on the Gaussian mutation and the random mutation. This algorithm was applied to the four classical test functions of 100,200 and 400 dimension and simulation shows that a marked improvement in performance over the traditional PSO.
出处
《计算机与数字工程》
2009年第7期13-16,共4页
Computer & Digital Engineering
关键词
微粒群算法
全局优化
变异算子
particle swarm optimization, global optimization, mutation operator
