基于支持向量分类器PCA方法及其在过程监控和故障检测中的应用  

Support Vector Classifier Based PCA with Application to Process

作　　者：陈国金[1] 梁军[1] 钱积新[1] 

机构地区：[1]浙江大学控制科学与工程学系,杭州310027

出　　处：《仪器仪表学报》2005年第1期54-58,共5页Chinese Journal of Scientific Instrument

基　　金：国家 973计划 (2 0 0 2 CB3 12 2 0 0 )资助项目

摘　　要：提出了基于支持向量分类器对过程进行性能监控和故障检测的改进 PCA方法 ,该方法避免了多元统计过程控制(MSPC)假设主元必须服从正态分布的前提。此外 ,通过对苯 -甲苯两组分精馏分离过程的仿真研究表明 ,该方法是有效的 ,并具有比传统多元统计过程控制更为优越的性能。Multivariate statistical process control (MSPC) has successfully been applied to performance monitoring and fault detection for industrial processes. However, classical methods of MSPC are based on the premise that separated latent variable must be subjected to normal distribution, which can't often hold in practical processes. A new PCA based on support vector classifier to performance monitoring and fault detection for process is presented, which can also be used for process which have non-Gaussian latent variable to monitor. Furthermore, a binary distillation column for benzene-toluene is monitored and detected by the present method, the simulation result verifies the more effectiveness and better performance of the PCA based on support vector classifier than the classical PCA.

关 键 词：分类器 支持向量 性能监控 过程监控 故障检测 过程控制 仿真研究 多元统计 主元 正态分布 

分 类 号：TH137[机械工程—机械制造及自动化] TP183[自动化与计算机技术—控制理论与控制工程]