摘要
针对传统组合优化方法用于故障特征选择的缺陷问题,提出了基于人工免疫和混沌思想的混合粒子群优化算法的特征选择策略。引入混沌优化和人工免疫系统中的克隆选择机制,利用克隆和混沌变异等算子对算法进行改进,提高种群的多样性,增强了算法跳出局部极值的能力。实验结果表明,该混合粒子群算法比常规粒子群算法具有更快的优化速度,有效提高了特征选择效率,使故障诊断精度有所提高。
Traditional methods for combinatorial optimization defects for fault feature selection problem,Feature selection strategy is proposed based on artificial Immune and chaos hybrid particle swarm optimization algorithm.The introduction of chaos optimization and artificial immune system clonal selection mechanism,the use of cloning and chaotic mutation operator to improve algorithm to improve population diversity,and enhance algorithm ability to jump out of local optimum.Experimental results show that hybrid particle swarm algorithm than conventional particle swarm optimization algorithm has a faster speed,improve efficiency of feature selection,so that improved accuracy of fault diagnosis.
出处
《煤矿机械》
北大核心
2011年第11期244-247,共4页
Coal Mine Machinery
关键词
粒子群算法
混沌
免疫接种
特征选择
particle swarm optimization
chaos
immunization
feature selection
