摘要
针对非线性优化问题讨论一种基于混合信息的粒子群优化算法,该算法考虑了最优个体和最差个体获取信息,结合自适应变异算子确定下一步搜索方向。自适应变异依据适应值大小调整速度惯性因子、改变搜索方向。仿真实验结果表明,新的算法收敛,具有很高的搜索效率和求解精度。
A new Particle Swarm Optimization（PSO） arithmetic based on hybrid information is presented, which covers the advantages to get available information from the best individual and the worst individual. Adaptive mutation arithmetic is also used to adjust the searching direction of nonlinear function problem, in which the speed weight ratio is mutated according to fitness of the objective function. Simulation results show that the nonlinear function problems can be solved with greater searching efficiency and better solution accuracy.
出处
《计算机工程》
CAS
CSCD
北大核心
2008年第7期176-177,183,共3页
Computer Engineering
基金
湖南省自然科学基金资助项目(06JJ5112)
湘潭大学学校跨学科交叉基金资助项目(05IND04)
关键词
混合信息
粒子群优化
自适应变异
非线性优化
hybrid information
Particle Swarm Optimization（PSO）
adaptive mutation
nonlinear optimization
