|Table of Contents|

Semi-supervised Sparse Discriminant Kernel Locally LinearEmbedding for Nonlinear Process Fault Detection(PDF)

《南京师大学报(自然科学版)》[ISSN:1001-4616/CN:32-1239/N]

Issue:
2018年04期
Page:
49-
Research Field:
·数学与计算机科学·
Publishing date:

Info

Title:
Semi-supervised Sparse Discriminant Kernel Locally LinearEmbedding for Nonlinear Process Fault Detection
Author(s):
Ren Shijin1Li Xinyu2Xu Guiyun2Pan Jianhan1Yang Maoyun1
(1.School of Computer Science & Technology,Jiangsu Normal University,Xuzhou 221116,China)(2.School of Mechatronic Engineering,China University of Mining and Technology,Xuzhou 221116,China)
Keywords:
process fault detectionkernel locally linear embeddingsemi-supervised learningFisher discriminant analysissparse representation
PACS:
TP391.9
DOI:
10.3969/j.issn.1001-4616.2018.04.009
Abstract:
Complex processes usually work under harzardous environment and varying operation conditions,and the process data exhibits high nonlinearity,randomness and local manifold structure. In recent years,kernel locally linear embedding(KLLE)has been successfully applied in fault detection for complex processes. However,KLLE is an unsupervised learning method and can preserve the nonlinear locality of the data while ignoring the global/nonlocal discriminant information. To address the issue,a semi-supervised sparse discriminant kernel locally linear embedding(SSDKLLE)approach is developed and applied to fault detection for nonlinear process in this work. The main contributions of the developed algorithm are summed as follows:(1)Exploiting the labeled and unlabeled data samples,semi-supervised learning and Fisher discriminant analysis technique are introduced to KLLE,effectively revealing the global geometric discriminant information hidden in the original data,and the performance of the proposed method is thus enhanced;(2)Considering that a signal can be sparsely represented by a set of atoms through an optimization algorithm,sparse representation is introduced to determine the neighborhood of the samples,improving the robustness and locality preserving of the proposed approach;(3)Local neighborhood processing strategy and kernel trick are introduced to LLE to reduce nonlinearity and distributions of process data with the changing operations,enhancing the performances of multimode and nonlinear process monitoring approaches. Experimental result on TE simulation platform demonstrates the efficiency and effectiveness of the proposed algorithm.

References:

[1] WANG L,SHI H B. Improved kernel PLS-based fault detection approach for nonlinear chemical processes[J]. Chinese journal of chemical engineering,2014,22(6):657-663.
[2]TONG C D,PALAZOGLU A,YAN X F. Improved ICA for process monitoring based on ensemble learning and Bayesian inference[J]. Chemometrics and intelligent laboratory systems,2014,135:141-149
[3]KANO M,HASEBE S,HASHIMOTO I H. A new multivariate statistical process monitoring method using principal component analysis[J]. Computer & chemical engineering,2001,25(7/8):1103-1113.
[4]NOMIKOS P,MACGREGOR J F. Monitoring batch processes using multiway principal component analysis[J]. American chemical engineering journal,1994,40(8):1361-1375.
[5]GE Z,YANG C,SONG Z. Improved kernel PCA-based monitoring approach for nonlinear processes[J]. Chemical engineering science,2009,64(9):2245-2255.
[6]JIE Y. A nonlinear kernel Gaussian mixture model based inferential monitoring approach for fault detection and diagnosis of chemical processes[J]. Chemical engineering science,2012,68(1):506-519.
[7]JIANG Q C,YAN X F. Monitoring multi-mode plant-wide processes by using mutual information-based multi-block PCA,joint probability,and Bayesian inference[J]. Chemometrics and intelligent laboratory systems,2014,136:121-137.
[8]NOR N M,HUSSAIN M A,HASSAN C R. Fault diagnosis and classification framework using multi-scale classification based on kernel Fisher discriminant analysis for chemical process system[J]. Applied soft computing,2017,61:959-972.
[9]LIU Y J,CHEN T,YAO Y. Nonlinear process monitoring and fault isolation using extended maximum variance unfolding[J]. Journal of process control,2014,24(6):880-891.
[10]ZHANG Y W,FU Y J,WANG Z B,et al. Fault detection based on modified kernel semi-supervised locally linear embedding[J]. IEEE access,2018,6:479-487.
[11]RONG G,LIU S Y,SHAO J D. Fault diagnosis by locality preserving discriminant analysis and its kernel variation[J]. Computer & chemical engineering,2013,49:105-113.
[12]CHEN G,LIU F L,HUANG W. Sparse discriminant manifold projections for bearing fault diagnosis[J]. Journal of sound and vibration,2017,399:330-344.
[13]YU J B. Hidden Markov models combining local and global information for nonlinear and multimodal process monitoring[J]. Journal of process control,2010,20:344-359.
[14]WANG X G,FENG H C,FAN Y P. Fault detection and classification for complex processes using semi-supervised learning algorithm[J]. Chemometrics and intelligent laboratory systems,2015,149:24-32.
[15]ZHEN J H,SONG Z H. Semi-supervised learning for probabilistic partial least squares regression model and soft sensor application[J]. Journal of process control,2018,64:123-131.
[16]GE Z Q,ZHONG S Y,ZHANG Y W. Semi-supervised kernel learning for FDA model and its application for fault classification in industrial processes[J]. IEEE transactions on industrial informatics,2016,12(4):1403-1411.
[17]FENG J,WANG J,ZHANG H G,et al. Fault diagnosis method of joint fisher discriminant analysis based on the local and global manifold learning and its kernel version[J]. IEEE transactions on automation science and engineering,2016,13(1):122-133.
[18]WEI J,MENG M,WANG J B,et al. Adaptive semi-supervised dimensionality reduction with sparse representation using pairwise constraints[J]. Neurocomputing,2016,177:564-571.
[19]XU Y,SHEN F M,XU X,et al. Large-scale image retrieval with upervised sparse hashing[J]. Neurocomputing,2017,229:45-53.
[20]SUN R B,YANG Z B,CHEN X F,et al. Gear fault diagnosis based on the structured sparsity time-frequency analysis[J]. Mechanical systems and signal processing,2018,102:346-363.
[21]XIAO Z H,WANG H G,ZHOU J W. Robust dynamic process monitoring based on sparse representation preserving embedding[J]. Journal of process control,2016,40:119-133.
[22]GU J,JIAO L C,YANG S Y,et al. Sparse learning based fuzzy c-means clustering[J]. Knowledge-based systems,2017,119:113-125.
[23]ZHANG Y,XIANG M,YANG B. Linear dimensionality reduction based on hybrid structure preserving projections[J]. Neurocomputing,2016,173:518-529.
[24]SHAO Z F,ZHANG L. Sparse dimensionality reduction of hyperspectral image based on semi-supervised local Fisher discriminant analysis[J]. International journal of applied earth observation and geoinformation,2014,31:122-129.
[25]马小虎,谭延琪. 基于鉴别稀疏保持嵌入的人脸识别算法[J]. 自动化学报,2014,40(1):73-82.
[26]CHEN P,JIAO L C,LIU F,et al. Semi-supervised double sparse graphs based discriminant analysis for dimensionality reduction[J]. Pattern recognition,2017,61:361-378.
[27]HE G,DING K,LIN H. Gearbox coupling modulation separation method based on match pursuit and correlation filtering[J]. Mechanical systems and signal processing,2016,66/67:597-611.
[28]李海山. 基于稀疏表示理论的地震信号处理方法研究[D]. 青岛:中国石油大学,2013.
[29]杜佳兵,唐刚,王华庆. 基于信号空间压缩感知算法的机械故障诊断[J]. 北京化工大学学报(自然科学版),2017,44(5):85-90.
[30]VAREWYCK M,MARTENS J P. A practical approach to model selection for support vector machines with a Gaussian kernel[J]. IEEE transactions on systems,man,and cybernetics Part B(Cybernetics),2011,41(2):330-340.
[31]郑鑫,田学民,张汉元. 基于动态稀疏保局投影的故障检测方法[J]. 化工学报,2016,67(3):833-838.
[32]YU L,YANF D,WANG H. Sparse multiple maximum scatter difference for dimensionality reduction[J]. Digital signal processing,2017,62,91-100.
[33]DOWNS J J,VOGEL E F. A plant-wide industrial process control problem[J]. Computers & chemical engineering,1993,17(3):245-255.
[34]LEE J M,QIN S J,LEE I B. Fault detection and diagnosis based on modified independent component analysis[J]. American chemical engineering journal,2006,52(10):3501-3514.
[35]文巧钧. 基于状态空间模型的复杂动态过程监测方法研究[D]. 杭州:浙江大学,2014.

Memo

Memo:
-
Last Update: 2018-12-30