Research on Feature Selection based on Improved Particle Swarm Optimization
Feature selection is one of key technologies for fault detection.Especially for high dimensional data,Feature selection can not only find the feature subset with sufficient information,but also improve the classification accuracy and efficiency.In order to decrease the number of detection parameter in fault detection of one equipment,the paper proposes one feature selection method based on improved particle swarm optimization,the method applies the quantum evolution thoughts to PSO.Firstly,the particle is restricted in the range from-π/2 to 0,so the particle can correspond to the quantum angle.Secondly,the parameter optimization function of feature selection is designed according to clustering criterion.Thirdly,the algorithm flow is designed.In fault detection of one equipment,the improved algorithm can decrease the number of parameter from 25 to 6.
Feature Selection Particle Optimization Swarm Quantum Evolution Fault Detection
Guo Qing Wang Jun Bo Jia Xu Yuan Li
150 MailBox,Baoji,Shanxi,China
国际会议
广州
英文
2651-2654
2012-11-16（万方平台首次上网日期，不代表论文的发表时间）