Condition monitoring of rotating machines under time-varying conditions based on adaptive canonical variate analysis

Abstract Condition monitoring signals obtained from rotating machines often demonstrate a highly non-stationary and transient nature due to internal natural deterioration characteristics of their constituent components and external time-varying operational conditions. Traditional multivariate statistical monitoring approaches are based on the assumption that the underlying processes are linear and static and are apt to interpret the normal changes in operating conditions as faults, which would result in high false positive rates. On the other hand, the development of robust diagnostic techniques for the detection of incipient faults remains a challenge for researchers, given the difficulty of finding an appropriate trade-off between a low false positive ratio and early detection of emerging faults. To address these issues, this paper proposes a novel adaptive fault detection approach based on the canonical residuals (CR) induced by the combination of canonical variate analysis (CVA) and matrix perturbation theory for the monitoring of dynamic processes where variations in operating conditions are incurred. The canonical residuals are calculated based upon the distinctions between past and future measurements and are able to effectively detect emerging faults while still maintaining a low false positive rate. The effectiveness of the developed diagnostic model for the detection of abnormalities in industrial processes was demonstrated for slow involving faults in case studies of two operational industrial high-pressure pumps. In comparison with the variable-based and canonical correlation-based statistical monitoring approaches, the proposed canonical residuals-adaptive canonical variate analysis (CR-ACVA) fault detection method has demonstrated its superiorities by the detailed performance comparisons.

[1]  Xiaochuan Li,et al.  Multidimensional prognostics for rotating machinery: A review , 2017 .

[2]  Horst Bischof,et al.  Appearance models based on kernel canonical correlation analysis , 2003, Pattern Recognit..

[3]  S. Joe Qin,et al.  Consistent dynamic PCA based on errors-in-variables subspace identification , 2001 .

[4]  Andrew D. Ball,et al.  An approach to fault diagnosis of reciprocating compressor valves using Teager-Kaiser energy operator and deep belief networks , 2014, Expert Syst. Appl..

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

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

[7]  A. Ben Hamza,et al.  Dynamic independent component analysis approach for fault detection and diagnosis , 2010, Expert Syst. Appl..

[8]  Weihua Li,et al.  Recursive PCA for adaptive process monitoring , 1999 .

[9]  Jin Hyun Park,et al.  Fault detection and identification of nonlinear processes based on kernel PCA , 2005 .

[10]  A. Negiz,et al.  Statistical monitoring of multivariable dynamic processes with state-space models , 1997 .

[11]  Cristobal Ruiz-Carcel,et al.  Canonical Variate Analysis for Performance Degradation under Faulty Conditions , 2016 .

[12]  H. Riedwyl,et al.  Standard Distance in Univariate and Multivariate Analysis , 1986 .

[13]  Kevin Judd,et al.  Modelling the dynamics of nonlinear time series using canonical variate analysis , 2002 .

[14]  R. Braatz,et al.  Fault detection of process correlation structure using canonical variate analysis-based correlation features , 2017 .

[15]  Jicong Fan,et al.  Fault detection and diagnosis of non-linear non-Gaussian dynamic processes using kernel dynamic independent component analysis , 2014, Inf. Sci..

[16]  Ahmed Hafaifa,et al.  Vibrations supervision in gas turbine based on parity space approach to increasing efficiency , 2015 .

[17]  Steven X. Ding,et al.  Recursive identification algorithms to design fault detection systems , 2010 .

[18]  Steven X. Ding,et al.  Fault Detection for Non-Gaussian Processes Using Generalized Canonical Correlation Analysis and Randomized Algorithms , 2018, IEEE Transactions on Industrial Electronics.

[19]  Yi Cao,et al.  Canonical Variate Dissimilarity Analysis for Process Incipient Fault Detection , 2018, IEEE Transactions on Industrial Informatics.

[20]  Edwin Lughofer,et al.  Fault detection in multi-sensor networks based on multivariate time-series models and orthogonal transformations , 2014, Inf. Fusion.

[21]  Donghua Zhou,et al.  Incipient fault detection with smoothing techniques in statistical process monitoring , 2017 .

[22]  David,et al.  Canonical Variable Analysis and Long Short-term Memory for Fault Diagnosis and Performance Estimation of a Centrifugal Compressor , 2018 .

[23]  Jie Zhang,et al.  Fault detection in dynamic processes using a simplified monitoring-specific CVA state space modelling approach , 2012, Comput. Chem. Eng..

[24]  Tricia J. Willink.  Efficient Adaptive SVD Algorithm for MIMO Applications , 2008, IEEE Transactions on Signal Processing.

[25]  Okyay Kaynak,et al.  Improved PLS Focused on Key-Performance-Indicator-Related Fault Diagnosis , 2015, IEEE Transactions on Industrial Electronics.

[26]  Asoke K. Nandi,et al.  FAULT DETECTION USING SUPPORT VECTOR MACHINES AND ARTIFICIAL NEURAL NETWORKS, AUGMENTED BY GENETIC ALGORITHMS , 2002 .

[27]  Lamiaa M. Elshenawy,et al.  Recursive Fault Detection and Isolation Approaches of Time-Varying Processes , 2012 .

[28]  Yingwei Zhang,et al.  Efficient recursive canonical variate analysis approach for monitoring time‐varying processes , 2017 .

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

[30]  Sondipon Adhikari,et al.  Rayleigh Quotient and Dissipative Systems , 2008 .