Plant-wide fault detection using graph signal processing

This paper introduces a graph signal processing (GSP) framework for industrial process monitoring. Through a graph built to represent the variables of an industrial plant and their relationships, we show how to transform each sample vector observation using the graph Fourier transform (GFT) in order to infer the status of the plant. To validate the proposed method we compare it to the state of the art PCA T2 and Q methods for the Tennessee Eastman Process (TEP), a widely used benchmark for validation of methods in control and process monitoring. The results show our method outperforms PCA in most cases.

[1]  Nina F. Thornhill,et al.  Cause-and-effect analysis in chemical processes utilizing XML, plant connectivity and quantitative process history , 2009, Comput. Chem. Eng..

[2]  E. F. Vogel,et al.  A plant-wide industrial process control problem , 1993 .

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

[4]  Sunil K. Narang,et al.  Compact Support Biorthogonal Wavelet Filterbanks for Arbitrary Undirected Graphs , 2012, IEEE Transactions on Signal Processing.

[5]  Ping Zhang,et al.  A comparison study of basic data-driven fault diagnosis and process monitoring methods on the benchmark Tennessee Eastman process , 2012 .

[6]  Fan Yang,et al.  Signed Directed Graph modeling of industrial processes and their validation by data-based methods , 2010, 2010 Conference on Control and Fault-Tolerant Systems (SysTol).

[7]  Jelena Kovacevic,et al.  Discrete Signal Processing on Graphs: Sampling Theory , 2015, IEEE Transactions on Signal Processing.

[8]  Adolfo Crespo Marquez,et al.  A framework for warranty management in industrial assets , 2012, Comput. Ind..

[9]  Antonio Ortega,et al.  Towards a sampling theorem for signals on arbitrary graphs , 2014, 2014 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP).

[10]  Pascal Frossard,et al.  The emerging field of signal processing on graphs: Extending high-dimensional data analysis to networks and other irregular domains , 2012, IEEE Signal Processing Magazine.

[11]  Luiz Affonso Guedes,et al.  Economic evaluation of industrial processes with ontology , 2014, Proceedings of the 11th IEEE International Conference on Networking, Sensing and Control.

[12]  Pierre Vandergheynst,et al.  A windowed graph Fourier transform , 2012, 2012 IEEE Statistical Signal Processing Workshop (SSP).

[13]  Wenyou Du,et al.  Process Fault Detection Using Directional Kernel Partial Least Squares , 2015 .

[14]  Nina F. Thornhill,et al.  Automatic detection and quantification of stiction in control valves , 2006 .

[15]  Jian Wang,et al.  Fault Diagnosis Method of Joint Fisher Discriminant Analysis Based on the Local and Global Manifold Learning and Its Kernel Version , 2016, IEEE Transactions on Automation Science and Engineering.

[16]  Yi Cao,et al.  A benchmark of Canonical Variate Analysis for fault detection and diagnosis , 2014, 2014 UKACC International Conference on Control (CONTROL).

[17]  Zhi-Huan Song,et al.  A novel fault diagnosis system using pattern classification on kernel FDA subspace , 2011, Expert Syst. Appl..

[18]  José M. F. Moura,et al.  Discrete Signal Processing on Graphs , 2012, IEEE Transactions on Signal Processing.

[19]  Luiz Affonso Guedes,et al.  Ontology for computer-aided fault tree synthesis , 2014, Proceedings of the 2014 IEEE Emerging Technology and Factory Automation (ETFA).

[20]  Fan Yang,et al.  Qualitative Fault Detection and Hazard Analysis Based on Signed Directed Graphs for Large-Scale Complex Systems , 2010 .

[21]  Nina F. Thornhill,et al.  A practical method for identifying the propagation path of plant-wide disturbances , 2008 .

[22]  Sunil K. Narang,et al.  Perfect Reconstruction Two-Channel Wavelet Filter Banks for Graph Structured Data , 2011, IEEE Transactions on Signal Processing.

[23]  Thomas F. Edgar,et al.  Process Dynamics and Control , 1989 .

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

[25]  José M. F. Moura,et al.  Discrete signal processing on graphs: Graph fourier transform , 2013, 2013 IEEE International Conference on Acoustics, Speech and Signal Processing.

[26]  Nina F. Thornhill,et al.  Automatic generation of bond graph models of process plants , 2008, 2008 IEEE International Conference on Emerging Technologies and Factory Automation.

[27]  José M. F. Moura,et al.  Discrete signal processing on graphs: Graph filters , 2013, 2013 IEEE International Conference on Acoustics, Speech and Signal Processing.

[28]  Pierre Vandergheynst,et al.  Wavelets on Graphs via Spectral Graph Theory , 2009, ArXiv.

[29]  Antonio Ortega,et al.  An optimization framework for combining multiple graphs , 2016, 2016 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP).

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

[31]  Mohieddine Jelali,et al.  Revision of the Tennessee Eastman Process Model , 2015 .

[32]  Nina F. Thornhill,et al.  Advances and new directions in plant-wide disturbance detection and diagnosis , 2007 .

[33]  Xuefeng Yan,et al.  Plant-wide process monitoring based on mutual information-multiblock principal component analysis. , 2014, ISA transactions.