基于多尺度主元分析方法的统计过程监视

基于主元分析和小波变换结合的基本理论，对Bakshi提出的MSPCA算法进行改进，提出一种新的多尺度主元分析方法（MSPCA）．MSPCA应用小波变换将每个变量信号依次分解成逼近系数和多个尺度的细节系数，把各个尺度的系数聚集在单独的矩阵中，在各个尺度建立相应的PCA模型，进行多尺度过程监视．针对TE过程的两种干扰，分别应用PCA和MSPCA两种方法做仿真试验．仿真实验结果表明：与PCA相比，MSPCA能有效地检测和识别过程中不同频率故障，减少误报警，提高了过程监视的可靠性．

Statistical Process Monitoring Based on Multi-scale Principal Component Analysis

Based on the theory of combining principal component analysis with wavelet transforms,a new multi-scale principal component analysis is proposed,which is different from the MSPCA formulation Bakshi developed.Using wavelet transforms,the individual variable signals are decomposed into approximations and details at different scales.Coefficients from each scale are collected in separate matrices,and a PCA model is then constructed to extract correlation at each scale.The results of simulation on TE process demonstrate: comparing with principal component analysis,multi-scale principal component analysis can be used to effectively detect different resolution variation,decrease false alarms,and increase the reliability of process monitoring.