PSO与LS混合算法在故障特征选择中的应用  

A Hybrid Algorithm Combined Particle Swarm Optimization and Laplace Score in Fault Feature Selection

作　　者：张亚龙[1] 赵荣珍[1] 王雪冬[1] 张娟[1] 

机构地区：[1]兰州理工大学机电工程学院,兰州730050

出　　处：《机械设计与研究》2016年第4期43-45,49,共4页Machine Design And Research

基　　金：高等学校博士学科点专项科研基金资助项目(20136201110004)

摘　　要：为了提高分类器的分类精度和泛化能力,提出一种基于粒子群算法(Particle Swarm Optimization,PSO)和拉普拉斯分值(Laplace Score,LS)的混合式故障特征选择方法。该方法首先采用过滤式的特征选择方法(LS)对原始特征集进行筛选,然后利用PSO在经过精简的特征子空间里进行随机搜索,搜索过程中以支持向量机的分类准确率为适应度函数,选择出最优特征子集。用双跨转子实验台的一组故障特征数据集进行验证。实验结果表明,该方法可以有效地筛选出规模较小且最有辨别力的特征子集,能显著提高分类器的分类准确率及效率。In order to improve the accuracy and generalization capability of classifier, the fault feature selection method based on particle swarm optimization algorithm and laplace score was proposed. In this method first laplace score was used to select features, then the optimal feature subset on the compact feature subset was searched by particle swarm optimization algorithm, in the search process used the accuracy of support vector machine as fitness function. A falut feature date set for a double-span rotor system was taken as a study objet. The results show that this method can filtrate a most discrimination classification feature set and improve the accuracy and efficiency of the classifier.

关 键 词：特征选择 特征子集 拉普拉斯分值 粒子群算法 支持向量机 

分 类 号：TH165[机械工程—机械制造及自动化] TP18[自动化与计算机技术—控制理论与控制工程]