Weighted principal component analysis applied to continuous stirred tank reactor system with time-varying

Fault detection approach based on principal component analysis (PCA) has been applied widely and effectively to chemical processes monitoring system. A novel PCA approach called weighted PCA (WPCA) for continuous stirred tank reactor system (CSTR) with time-varying is addressed in the paper. Time-varying can cause unfavorable influence on feature extraction, but weighted PCA approach can obtain slow features information of observed data in CSTR with time-varying. The monitoring statistical indices are based on WPCA approach and their confidence limits are computed by kernel density estimation (KDE). A simulation illustrated that the proposed method achieves better dynamical performance from the perspective of fault detection rate and fault detection time than PCA approach.

[1]  Si-Zhao Joe Qin,et al.  Survey on data-driven industrial process monitoring and diagnosis , 2012, Annu. Rev. Control..

[2]  Ying-wei Zhang,et al.  Improved multi-scale kernel principal component analysis and its application for fault detection , 2012 .

[3]  Xuefeng Yan,et al.  Chemical processes monitoring based on weighted principal component analysis and its application , 2012 .

[4]  Jicong Fan,et al.  Online monitoring of nonlinear multivariate industrial processes using filtering KICA–PCA , 2014 .

[5]  Bo Du,et al.  Slow Feature Analysis for Change Detection in Multispectral Imagery , 2014, IEEE Transactions on Geoscience and Remote Sensing.

[6]  Laurenz Wiskott,et al.  Independent Slow Feature Analysis and Nonlinear Blind Source Separation , 2004, ICA.

[7]  Yingwei Zhang,et al.  Enhanced statistical analysis of nonlinear processes using KPCA, KICA and SVM , 2009 .

[8]  Xuemin Tian,et al.  Sparse Kernel Locality Preserving Projection and Its Application in Nonlinear Process Fault Detection , 2013 .

[9]  Tao Chen,et al.  Nonlinear process monitoring and fault isolation using extended maximum variance unfolding , 2014 .

[10]  Fuli Wang,et al.  Reconstruction in integrating fault spaces for fault identification with kernel independent component analysis , 2013 .

[11]  Shuai Li,et al.  Dynamic processes monitoring using recursive kernel principal component analysis , 2012 .

[12]  Xue-feng Yan,et al.  Weighted kernel principal component analysis based on probability density estimation and moving window and its application in nonlinear chemical process monitoring , 2013 .

[13]  José M. Rodríguez,et al.  GPCA vs. PCA in Recognition and 3-D Localization of Ultrasound Reflectors , 2010, Sensors.

[14]  Terrence J. Sejnowski,et al.  Slow Feature Analysis: Unsupervised Learning of Invariances , 2002, Neural Computation.

[15]  Steven X. Ding,et al.  Data-driven design of monitoring and diagnosis systems for dynamic processes: A review of subspace technique based schemes and some recent results , 2014 .

[16]  ChangKyoo Yoo,et al.  Fault detection of batch processes using multiway kernel principal component analysis , 2004, Comput. Chem. Eng..