Canonical Variate Analysis for Performance Degradation under Faulty Conditions

Condition monitoring of industrial processes can minimize maintenance and operating costs while increasing the process safety and enhancing the quality of the product. In order to achieve these goals it is necessary not only to detect and diagnose process faults, but also to react to them by scheduling the maintenance and production according to the condition of the process. The objective of this investigation is to test the capabilities of canonical variate analysis (CVA) to estimate performance degradation and predict the behaviour of a system affected by faults. Process data was acquired from a large-scale experimental multiphase flow facility operated under changing operational conditions where process faults were seeded. The results suggest that CVA can be used effectively to evaluate how faults affect the process variables in comparison to normal operation. The method also predicted future process behaviour after the appearance of faults, modelling the system using data collected during the early stages of degradation.

[1]  Jialin Liu,et al.  Multiple Sensor Fault Isolation Using Contribution Plots without Smearing Effect to Non-Faulty Variables , 2012 .

[2]  Furong Gao,et al.  Review of Recent Research on Data-Based Process Monitoring , 2013 .

[3]  Richard D. Braatz,et al.  Fault detection in industrial processes using canonical variate analysis and dynamic principal component analysis , 2000 .

[4]  Ali Cinar,et al.  Monitoring of multivariable dynamic processes and sensor auditing , 1998 .

[5]  Pan Hongxia,et al.  Study of fault feature extraction based on KPCA optimized by PSO algorithm , 2010, FUZZ-IEEE.

[6]  Okyay Kaynak,et al.  An LWPR-Based Data-Driven Fault Detection Approach for Nonlinear Process Monitoring , 2014, IEEE Transactions on Industrial Informatics.

[7]  C. Yoo,et al.  Nonlinear process monitoring using kernel principal component analysis , 2004 .

[8]  Joan Colomer-Llinas,et al.  Fault Diagnosis of Batch Processes Release using PCA Contribution Plots as Fault Signatures , 2011, ICEIS.

[9]  Wallace E. Larimore,et al.  Canonical variate analysis in identification, filtering, and adaptive control , 1990, 29th IEEE Conference on Decision and Control.

[10]  Yi Cao,et al.  Nonlinear Dynamic Process Monitoring Using Canonical Variate Analysis and Kernel Density Estimations , 2010, IEEE Transactions on Industrial Informatics.

[11]  Richard D. Braatz,et al.  Fault Detection and Diagnosis in Industrial Systems , 2001 .

[12]  Dale E. Seborg,et al.  Identification of Multivariable, Linear, Dynamic Models: Comparing Regression and Subspace Techniques , 2002 .

[13]  Xiaoguang Chen,et al.  Research on fault identification for complex system based on generalized linear canonical correlation analysis , 2012, 2012 IEEE International Conference on Automation Science and Engineering (CASE).

[14]  Dale E. Seborg,et al.  Fault Detection Using Canonical Variate Analysis , 2004 .

[15]  Ru Qing Chen.  Advances in Data-Driven Monitoring Methods for Complex Process , 2013 .

[16]  Hong Zhou,et al.  Decentralized Fault Diagnosis of Large-Scale Processes Using Multiblock Kernel Partial Least Squares , 2010, IEEE Transactions on Industrial Informatics.

[17]  Dale E. Seborg,et al.  Identification of the Tennessee Eastman challenge process with subspace methods , 2000 .

[18]  Cristobal Ruiz-Carcel,et al.  Statistical process monitoring of a multiphase flow facility , 2015 .

[19]  H. J. Borsje.  Fault detection in boilers using canonical variate analysis , 1999, Proceedings of the 1999 American Control Conference (Cat. No. 99CH36251).

[20]  Dale E. Seborg,et al.  Identification of chemical processes using canonical variate analysis , 1994 .

[21]  Xuemin Tian,et al.  A new fault isolation method based on unified contribution plots , 2011, Proceedings of the 30th Chinese Control Conference.

[22]  Xiaobo Chen,et al.  Multivariate industrial process monitoring based on the integration method of canonical variate analysis and independent component analysis , 2012 .