A Nonlinear Dynamic Process Fault Detection Method Based on
 Kernel State Space Independent Component Analysis

Abstract

Abstract:

A fault detection method based on kernel state space independent component analysis (KSSICA) was proposed in this paper considering the nonlinear and dynamic characteristics of industrial processes. Kernel canonical variate analysis (KCVA) was adopted to project the nonlinear and dynamic process data into the kernel state space, and the state data which were uncorrelated were obtained. Based on the state data’s time structure matrix which is the weighted sum of the state data’s different timedelayed covariance matrices, an ICA statistical model was constructed to extract the independent component feature data from the state data, and the monitoring statistics were built to detect process faults. The fault detection results on the Tennessee Eastman benchmark process demonstrate that the proposed KSSICAbased fault detection method can detect the process faults more agilely and obtain a higher fault detection rate than the conventional fault detection method based on dynamic kernel principal component analysis (DKPCA).

Key words: fault detection； nonlinearity； dynamic characteristic； kernel canonical variate analysis； independent component analysis； fault detection rate

CLC Number:

TP 277

CAI Lianfang,TIAN Xuemin,ZHANG Ni. A Nonlinear Dynamic Process Fault Detection Method Based on

Kernel State Space Independent Component Analysis

[J]. Journal of Shanghai Jiaotong University, 2014, 48(07): 971-976.

References

［1］Zhang Y W, An J Y, Zhang H L. Monitoring of timevarying processes using kernel independent component analysis［J］. Chemical Engineering Science, 2013, 88: 2332.

［2］Stubbs S, Zhang J, Morris J. Fault detection in dynamic processes using a simplified monitoringspecific CVA state space modeling approach［J］. Computers and Chemical Engineering, 2012, 41(11): 7787.

［3］Lee J, Kang B, Kang S H. Integrating independent component analysis and local outlier factor for plantwide process monitoring［J］. Journal of Process Control, 2011, 21(7): 10111021.

［4］Yu J B. Local and global principal component analysis for process monitoring［J］. Journal of Process Control, 2012, 22(7): 13581373.

［5］文成林, 胡静, 王天真，等. 相对主元分析及其在数据压缩和故障诊断中的应用研究［J］. 自动化学报, 2008, 34(9): 11281139.

WEN Chenglin, HU Jing, WANG Tianzhen，et al. Relative PCA with applications of data compression and fault diagnosis［J］. Acta Automatica Sinica, 2008, 34(9): 11281139.

［6］Alcala C F, Qin S J. Reconstructionbased contribution for process monitoring with kernel principal component analysis［J］. Industrial & Engineering Chemistry Research, 2010, 49(17): 78497857.

［7］Russell E L, Chiang C H, Braatz R D. Fault detection in industrial processes using canonical variate analysis and dynamic principal component analysis［J］. Chemometrics and Intelligent Laboratory Systems, 2000, 51(1): 8193.

［8］Jia M X, Chu F, Wang F L, et al. Online batch process monitoring using batch dynamic kernel principal component analysis［J］. Chemometrics and Intelligent Laboratory Systems, 2010, 101(2): 110122.

［9］Negiz A, Cinar A. Monitoring of multivariable dynamic processes and sensor auditing［J］. Journal of Process Control, 1998, 8(56): 375380.

［10］Melzer T, Reiter M, Bischof H. Appearance models based on kernel canonical correlation analysis［J］. Pattern Recognition, 2003, 36(9): 19611971.

［11］Tian X M, Zhang X L, Deng X G, et al. Multiway kernel independent component analysis based on feature samples for batch process monitoring［J］. Neurocomputing, 2009, 72(7): 15841596.

［12］蔡连芳, 田学民, 张妮. 基于时序结构KICA和OCSVM的过程故障检测方法［J］. 清华大学学报：自然科学版, 2012, 52(9): 12051209.

CAI Lianfang, TIAN Xuemin, ZHANG Ni. Process fault detection method using timestructure KICA and OCSVM［J］.

Journal of Tsinghua University： Science and Technology, 2012, 52(9): 12051209.

［13］Chiang L H, Russell E L, Braatz R D. Fault detection and diagnosis in industrial systems［M］. London, Britain: SpringerVerlag, 2001.

［14］Lee J M, Yoo C K, Choi S W, et al. Nonlinear process monitoring using kernel principal component analysis［J］. Chemical Engineering Science, 2004, 59(1): 223234.

［15］Ku W, Storer R H, Georgakis C. Disturbance detection and isolation by dynamic principal component analysis［J］. Chemometrics and Intelligent Laboratory Systems, 1995, 30(1): 179196.

［16］Mahadevan S, Shah S L. Fault detection and diagnosis in process data using oneclass support vector machines［J］. Journal of Process Control, 2009, 19(10): 16271639.

[1]	HUANG Jian, YANG Xu. Online Weighted Slow Feature Analysis Based Fault Detection Algorithm [J]. Journal of Shanghai Jiao Tong University, 2020, 54(11): 1142-1150.
[2]	HU Jing1，WANG Chunxia2，WEN Chenglin2，LI Ping1. Quality Monitoring of Nonlinear Process Based on Kernel Projection to Quality Latent Structure [J]. Journal of Shanghai Jiaotong University, 2015, 49(06): 737-742.
[3]	WANG Limin1，2,ZHOU Donghua1,ZHU Chengjie2. Composite Iterative Learning FaultTolerant Guaranteed Cost Controller Design for Batch Process [J]. Journal of Shanghai Jiaotong University, 2015, 49(06): 743-750.
[4]	QIN Liguoa,HE Xiaoa,b,ZHOU Donghuaa,b. A Fault Estimation Method Based on Robust Residual Generators [J]. Journal of Shanghai Jiaotong University, 2015, 49(06): 768-774.
[5]	SAHNG Jun,CHEN Maoyin,ZHOU Donghua. Incipient Fault Detection Using Transformed Component Statistical Analysis [J]. Journal of Shanghai Jiaotong University, 2015, 49(06): 799-805.
[6]	CONG Ya,GE Zhiqiang,SONG Zhihuan. Multi-Rate Principle Component Analysis for Process Monitoring [J]. Journal of Shanghai Jiaotong University, 2015, 49(06): 762-767.
[7]	CHU Zhenzhong1,ZHU Daqi1,ZHANG Mingjun2. Terminal Sliding Mode Observer Based Fault Reconstruction for Underwater Vehicle Thruster [J]. Journal of Shanghai Jiaotong University, 2015, 49(06): 837-841.
[8]	LI Xiongjie1,2，ZHOU Donghua2. Fault Diagnosis Based on Particle Filter for Hybrid System [J]. Journal of Shanghai Jiaotong University, 2015, 49(06): 849-854.
[9]	ZHAO Xiaoqiang1,2,QIAN Junxiu1. A Fault Diagnosis Algorithm for Chemical Process Based on Dual-Kernel Independent Component Analysis [J]. Journal of Shanghai Jiaotong University, 2014, 48(07): 1004-1008.
[10]	WANG Tian-Zhen-1, 2 , GAO Di-Ju-1, LIU Ping-1, TANG Tian-Hao-1 . A Fault Detection Model Based on Dynamic Limit under Non-periodic Non-steady Conditions [J]. Journal of Shanghai Jiaotong University, 2012, 46(04): 607-612.
[11]	GUO Hongjie1,XU Chunling2,SHI Hongbo1. Multimode Process Monitoring Based on Local Neighborhood Standardization Strategy [J]. Journal of Shanghai Jiaotong University, 2015, 49(06): 868-875.

A Nonlinear Dynamic Process Fault Detection Method Based on
Kernel State Space Independent Component Analysis

Knowledge

Abstract

Cite this article

share this article

References

Related Articles 11

Recommended Articles

Metrics

Comments