Iterated Robust kernel Fuzzy Principal Component Analysis and application to fault detection

Abstract In this paper, we propose an Iterated Robust kernel Fuzzy Principal Component Analysis (IRkFPCA), which is the method that attempts to combine the advantages of the state of art methods and use a more accurate multi-objective function for jointly reducing the modeling errors, optimizing the robustness to outliers and improving the time complexity since it does not require the storage and inversion of the covariance matrix to obtain a memory-efficient approximation of kernel PCA. This proposed technique computes iteratively the principal components, which are used for modeling and fault detection. The detection stage is related to the evaluation of residuals, also known as detection indices, which are signals that reveal the fault presence. Those indices are obtained from the analysis of the difference between the process measurements and their estimations using the IRkFPCA technique. The performance of the proposed method is illustrated and compared to Iterated kernel Principal Component Analysis (IkPCA) and Iterated Principal Component Analysis (IPCA) methods through two simulated examples, one using synthetic data and the other using simulated continuously stirred tank reactor (CSTR) data. The results of the comparative studies reveal that the developed IRkFPCA method provides a better performance in terms of modeling and fault detection accuracies than the Iterated Robust Fuzzy Principal Component Analysis (IRFPCA) and Iterated kernel Principal Component Analysis (IkPCA) methods; while both methods provide improved accuracy over the Iterated Principal Component Analysis (IPCA) method.

[1]  Didier Maquin,et al.  Fault Detection and Isolation with Robust Principal Component Analysis , 2008, 2008 16th Mediterranean Conference on Control and Automation.

[2]  Érica C. M. Nascimento,et al.  Pharmacophoric Profile: Design of New Potential Drugs with PCA Analysis , 2012 .

[3]  P. Frank,et al.  Fault detection via factorization approach , 1990 .

[4]  Jin-Wook Chung,et al.  A Rule Based Approach to Network Fault and Security Diagnosis with Agent Collaboration , 2004, AIS.

[5]  Alan L. Yuille,et al.  Robust principal component analysis by self-organizing rules based on statistical physics approach , 1995, IEEE Trans. Neural Networks.

[6]  Celal Batur,et al.  Support vector machines for fault detection , 2002, Proceedings of the 41st IEEE Conference on Decision and Control, 2002..

[7]  Theodora Kourti,et al.  Statistical Process Control of Multivariate Processes , 1994 .

[8]  Gilles Mourot,et al.  An improved PCA scheme for sensor FDI: Application to an air quality monitoring network , 2006 .

[9]  S. V. N. Vishwanathan,et al.  Fast Iterative Kernel Principal Component Analysis , 2007, J. Mach. Learn. Res..

[10]  R. Patton,et al.  A Review of Parity Space Approaches to Fault Diagnosis , 1991 .

[11]  G. Parisi,et al.  Statistical Field Theory , 1988 .

[12]  B. Kannapiran,et al.  Artificial Neural Network Approach for Fault Detection in Pneumatic Valve in Cooler Water Spray System , 2010 .

[13]  José Ragot,et al.  Caractérisation des incertitudes dans un modèle linéaire statique , 2002 .

[14]  Rüdiger Wirth,et al.  Knowledge-based support of system analysis for the analysis of Failure modes and effects , 1996 .

[15]  Jianbo Yu,et al.  Fault Detection Using Principal Components-Based Gaussian Mixture Model for Semiconductor Manufacturing Processes , 2011, IEEE Transactions on Semiconductor Manufacturing.

[16]  V. Walton,et al.  Detecting Instrument Malfunctions in Control Systems , 1975, IEEE Transactions on Aerospace and Electronic Systems.

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

[18]  Pekka Teppola,et al.  Adaptive Fuzzy C-Means clustering in process monitoring , 1999 .

[19]  Qingsong Yang,et al.  MODEL-BASED AND DATA DRIVEN FAULT DIAGNOSIS METHODS WITH APPLICATIONS TO PROCESS MONITORING , 2004 .

[20]  Ferran Reverter,et al.  Kernel Methods for Dimensionality Reduction Applied to the «Omics» Data , 2012 .

[21]  Raghunathan Rengaswamy,et al.  A review of process fault detection and diagnosis: Part III: Process history based methods , 2003, Comput. Chem. Eng..

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

[23]  Naga Chaitanya Kavuri.  Data-Based Modeling: Application in Process Identification, Monitoring and Fault Detection , 2011 .

[24]  R. Blong,et al.  The 2003 Heat Wave in France: Dangerous Climate Change Here and Now , 2005, Risk analysis : an official publication of the Society for Risk Analysis.

[25]  Raghunathan Rengaswamy,et al.  A review of process fault detection and diagnosis: Part II: Qualitative models and search strategies , 2003, Comput. Chem. Eng..

[26]  J. A. Bandoni,et al.  Robust PCA and normal region in multivariate statistical process monitoring , 1996 .

[27]  Giancarlo Diana,et al.  Cross-validation methods in principal component analysis: A comparison , 2002 .

[28]  Yvon Tharrault,et al.  Diagnostic de fonctionnement par analyse en composantes principales (application à une station de traitement des eaux usées) , 2008 .

[29]  Michel Kinnaert,et al.  Fault diagnosis based on analytical models for linear and nonlinear systems - a tutorial , 2003 .

[30]  Stephen P. Boyd,et al.  Convex Optimization , 2004, Algorithms and Theory of Computation Handbook.

[31]  Steven W. Su,et al.  Online Support Vector Machine Applicationfor Model Based Fault Detection and Isolationof HVAC System , 2011 .

[32]  A. J. Morris,et al.  Multivariate Statistical Process Control in Chemicals Manufacturing , 1997 .

[33]  P. Luukka.  A New Nonlinear Fuzzy Robust PCA Algorithm and Similarity Classifier in Classification of Medical Data Sets , 2011 .

[34]  John E. Moody,et al.  Towards Faster Stochastic Gradient Search , 1991, NIPS.

[35]  H. Hotelling.  Analysis of a complex of statistical variables into principal components. , 1933 .

[36]  S. A. Vejtasa,et al.  An experimental study of steady state multiplicity and stability in an adiabatic stirred reactor , 1970 .

[37]  J. E. Jackson,et al.  Control Procedures for Residuals Associated With Principal Component Analysis , 1979 .

[38]  Eric J. Belasco,et al.  The Health Care Access Index as a Determinant of Delayed Cancer Detection Through Principal Component Analysis , 2012 .

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

[40]  David Zumoffen,et al.  From Large Chemical Plant Data to Fault Diagnosis Integrated to Decentralized Fault-Tolerant Control : Pulp Mill Process Application , 2008 .

[41]  J. Ragot,et al.  Analytical redundancy for systems with unknown inputs. Application to faults detection , 1993 .

[42]  Barry M. Wise,et al.  The process chemometrics approach to process monitoring and fault detection , 1995 .

[43]  S. Qin,et al.  Determining the number of principal components for best reconstruction , 2000 .

[44]  David M. Himmelblau,et al.  The possible cause and effect graphs (PCEG) model for fault diagnosis—I. Methodology , 1994 .

[45]  Mu Zhu,et al.  Automatic dimensionality selection from the scree plot via the use of profile likelihood , 2006, Comput. Stat. Data Anal..

[46]  S. Joe Qin,et al.  Statistical process monitoring: basics and beyond , 2003 .

[47]  Erkki Oja,et al.  Principal components, minor components, and linear neural networks , 1992, Neural Networks.

[48]  Donát Magyar,et al.  Application of the Principal Component Analysis to Disclose Factors Influencing on the Composition of Fungal Consortia Deteriorating Remained Fruit Stalks on Sour Cherry Trees , 2012 .

[49]  T. Severini.  Likelihood Methods in Statistics , 2001 .

[50]  Verron Sylvain,et al.  Fault Detection with Bayesian Network , 2008 .

[51]  Raghunathan Rengaswamy,et al.  A review of process fault detection and diagnosis: Part I: Quantitative model-based methods , 2003, Comput. Chem. Eng..

[52]  Paul D. Gader,et al.  2009 Special Issue: RKF-PCA: Robust kernel fuzzy PCA , 2009 .

[53]  A. Willsky,et al.  Analytical redundancy and the design of robust failure detection systems , 1984 .

[54]  Terence D. Sanger,et al.  Optimal unsupervised learning in a single-layer linear feedforward neural network , 1989, Neural Networks.

[55]  Hazem N. Nounou,et al.  State and parameter estimation for nonlinear biological phenomena modeled by S-systems , 2014, Digit. Signal Process..

[56]  C. Croux,et al.  Principal Component Analysis Based on Robust Estimators of the Covariance or Correlation Matrix: Influence Functions and Efficiencies , 2000 .

[57]  Zheng Chen,et al.  Fault Detection of Drinking Water Treatment Process Using PCA and Hotelling's T2 Chart , 2009 .

[58]  Bernhard Schölkopf,et al.  Iterative kernel principal component analysis for image modeling , 2005, IEEE Transactions on Pattern Analysis and Machine Intelligence.

[59]  Y. Pawitan.  In all likelihood : statistical modelling and inference using likelihood , 2002 .

[60]  S. Qin,et al.  Selection of the Number of Principal Components: The Variance of the Reconstruction Error Criterion with a Comparison to Other Methods† , 1999 .

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

[62]  Kamel Benothman,et al.  Diagnosis of uncertain linear systems: an interval approach , 2007 .

[63]  J. Golinval,et al.  Fault detection based on Kernel Principal Component Analysis , 2010 .

[64]  M. Nounou,et al.  Statistical Detection of Abnormal Ozone Levels Using Principal Component Analysis , 2012 .

[65]  Karim Mohammadi,et al.  Pattern Recognition and Fault Detection in MEMS , 2005, CORES.

[66]  Masayuki Tamura,et al.  A study on the number of principal components and sensitivity of fault detection using PCA , 2007, Comput. Chem. Eng..

[67]  I. Jolliffe.  Principal Component Analysis , 2002 .