Mixture Probabilistic Linear Discriminant Analyzer for Process Fault Classification

Fault misclassification results in misleading locations of root faults, which brings economic loss and safety concerns. In order to enhance the performance of fault classification, this paper proposes a novel mixture fault classifier based on probabilistic linear discriminant analyzer (PLDA). Unlike the probabilistic model with a single latent variable, more useful information can be extracted by PLDA using within- and between-class latent variables. With the introduced state and mixture variables, the proposed mixture PLDA (MPLDA) classifier assigns a test sample to different components for each class in probability. For the class label decision, a robust state inference strategy is developed, which includes investigating the effect of the shared between-class variable on conditional probability and adopting the voting scheme to collect evidences from training samples to identify the class source of the test sample. After classifier construction, the model parameters are obtained by the expectation-maximization (EM) algorithm. The validity of MPLDA classifier in fault classification is then illustrated by the application of the Tennessee Eastman (TE) process.

[1]  Zhiqiang Ge,et al.  Review on data-driven modeling and monitoring for plant-wide industrial processes , 2017 .

[2]  Eduardo Lleida,et al.  Estimation of the Number of Speakers with Variational Bayesian PLDA in the DIHARD Diarization Challenge , 2018, INTERSPEECH.

[3]  Steven X. Ding,et al.  A Review on Basic Data-Driven Approaches for Industrial Process Monitoring , 2014, IEEE Transactions on Industrial Electronics.

[4]  In-Beum Lee,et al.  Process monitoring based on probabilistic PCA , 2003 .

[5]  Jen-Tzung Chien,et al.  Mixture of PLDA for Noise Robust I-Vector Speaker Verification , 2016, IEEE/ACM Transactions on Audio, Speech, and Language Processing.

[6]  Changqing Liu,et al.  Modified partial least square for diagnosing key‐performance‐indicator‐related faults , 2018 .

[7]  Biao Huang,et al.  Two layered mixture Bayesian probabilistic PCA for dynamic process monitoring , 2017 .

[8]  Hans-Joachim Wunderlich,et al.  A Neural-Network-Based Fault Classifier , 2016, 2016 IEEE 25th Asian Test Symposium (ATS).

[9]  Feng Lin,et al.  Classification of non-tumorous skin pigmentation disorders using voting based probabilistic linear discriminant analysis , 2018, Comput. Biol. Medicine.

[10]  Zhiqiang Ge,et al.  Dynamic mixture probabilistic PCA classifier modeling and application for fault classification , 2015 .

[11]  A. Elayaperumal,et al.  Fault diagnostics of spur gear using decision tree and fuzzy classifier , 2017 .

[12]  Songhang Chen,et al.  An Improved Mixture of Probabilistic PCA for Nonlinear Data-Driven Process Monitoring , 2019, IEEE Transactions on Cybernetics.

[13]  Biao Huang,et al.  Process monitoring using a generalized probabilistic linear latent variable model , 2018, Autom..

[14]  Yingwei Zhang,et al.  Independent component analysis based on data‐driven reconstruction of multi‐fault diagnosis , 2017 .

[15]  Umar Mohammed,et al.  Probabilistic Models for Inference about Identity , 2012, IEEE Transactions on Pattern Analysis and Machine Intelligence.

[16]  Jialin Liu,et al.  Fault diagnosis using contribution plots without smearing effect on non-faulty variables , 2012 .

[17]  Yuan Yao,et al.  Reconstruction-Based Multivariate Process Fault Isolation Using Bayesian Lasso , 2018 .

[18]  Claude Delpha,et al.  Incipient fault detection and diagnosis based on Kullback-Leibler divergence using principal component analysis: Part II , 2015, Signal Process..

[19]  Christos Georgakis,et al.  Plant-wide control of the Tennessee Eastman problem , 1995 .

[20]  Lei Xie,et al.  Shrinking principal component analysis for enhanced process monitoring and fault isolation , 2013 .

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

[22]  D.G. Tzikas,et al.  The variational approximation for Bayesian inference , 2008, IEEE Signal Processing Magazine.

[23]  Mohammad Nizam Ibrahim,et al.  Support vector machine (SVM) for fault classification in radial distribution network , 2017 .

[24]  Zhiqiang Ge,et al.  HMM-Driven Robust Probabilistic Principal Component Analyzer for Dynamic Process Fault Classification , 2015, IEEE Transactions on Industrial Electronics.

[25]  James H. Elder,et al.  Probabilistic Linear Discriminant Analysis for Inferences About Identity , 2007, 2007 IEEE 11th International Conference on Computer Vision.

[26]  Sridha Sridharan,et al.  Improving PLDA speaker verification performance using domain mismatch compensation techniques , 2018, Comput. Speech Lang..

[27]  Richard D. Braatz,et al.  Perspectives on process monitoring of industrial systems , 2016, Annu. Rev. Control..

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

[29]  In-Beum Lee,et al.  Fault Detection Based on a Maximum-Likelihood Principal Component Analysis (PCA) Mixture , 2005 .

[30]  Zhiqiang Ge,et al.  Maximum-likelihood mixture factor analysis model and its application for process monitoring , 2010 .