文献类型：期刊文献

英文题名：Multi-block statistics local kernel principal component analysis algorithm and its application in nonlinear process fault detection

作者：Zhou, Bingqian[1];Gu, Xingsheng[1]

机构：[1]East China Univ Sci & Technol, Minist Educ, Key Lab Adv Control & Optimizat Chem Proc, Shanghai 200237, Peoples R China

年份：2020

卷号：376

起止页码：222

外文期刊名：NEUROCOMPUTING

收录：;EI(收录号：20194507632529);WOS:【SCI-EXPANDED(收录号:WOS:000503433800019)】；

基金：This work is supported by the National Natural Science Foundation of China (Grant No.61573144, 61773165, 61673175, 61973120), the Program of Introducing Talents of Discipline to Universities (the 111 Project) (Grant No. B17017), Fundamental Research Funds for the Central Universities (No. 222201917006).

语种：英文

外文关键词：Multi-block; Local KPCA; Statistics pattern analysis; Bayesian strategy analysis; Fault detection

摘要：It is vital for fault detection technology to extract features of industrial process data effectively. Local kernel principal component analysis (LKPCA) has proved its good performance in preserving global and local structural characteristics. However, it ignored useful high-order statistics of data, so multi-block statistics local kernel principal component analysis (MSLKPCA) algorithm integrating statistics pattern analysis (SPA) into LKPCA is proposed. The correlation coefficient matrix is first calculated and K-means clustering is adopted to divide the original variables into several blocks. Then the weighted SPA, which gives different weights to different samples in each window according to their distributions, is adopted to build statistic spaces containing both low-order and high-order statistics. After that, LKCPA is performed in each statistic space to realize feature extraction. To reduce the noise effect amplified by SPA, PCA is adopted in the residual space to remove noise. Bayesian strategy is used to fuse the results of each block and two monitoring statistics E-T and E-R are proposed to monitor the feature space and the residual space respectively. The Tennessee-Eastman (TE) process simulation shows the effectiveness and superiority of the proposed algorithm for process monitoring. (C) 2019 Elsevier B.V. All rights reserved.