Dynamic reconstruction based representation learning for multivariable process monitoring

Abstract Representation learning is a key step for fault detection in the process monitoring. With the consideration of time-correlated, this paper focus on developing a dynamic representation learning method yields high-order correlations in process monitoring. First, a general interpretation of AE net is presented based on Taylor expansion, which motivates the representation ability by the composition of multiple non-linear transformations. Due to the fact that the nearest neighbors over time are not necessarily the nearest spatial neighbors in dynamic process, a dynamic reconstruction is developed based on its k nearest neighbors. The reconstruction can not only maintain the separability, but also increase the distinguishable distance between categories. Finally, numerical results shown that a multi-layer Taylor network is efficiently to approximate the smooth sigmoid functions of AE net, and experiments on Tennessee Eastman process (TEP) illustrated the proposed method’s superior detectability, specifically for incipient faults.

[1]  Biao Huang,et al.  Survey on the theoretical research and engineering applications of multivariate statistics process monitoring algorithms: 2008-2017 , 2018, The Canadian Journal of Chemical Engineering.

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

[3]  John C. Young,et al.  Multivariate statistical process control with industrial applications , 2002, ASA-SIAM series on statistics and applied probability.

[4]  Nasser M. Nasrabadi,et al.  Pattern Recognition and Machine Learning , 2006, Technometrics.

[5]  Geoffrey E. Hinton,et al.  Reducing the Dimensionality of Data with Neural Networks , 2006, Science.

[6]  Chenglin Wen,et al.  Higher‐order correlation–based multivariate statistical process monitoring , 2018 .

[7]  Larry D. Hostetler,et al.  k-nearest-neighbor Bayes-risk estimation , 1975, IEEE Trans. Inf. Theory.

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

[9]  Chenglin Wen,et al.  Weighted time series fault diagnosis based on a stacked sparse autoencoder , 2017 .

[10]  Pascal Vincent,et al.  Representation Learning: A Review and New Perspectives , 2012, IEEE Transactions on Pattern Analysis and Machine Intelligence.

[11]  Rick L. Edgeman,et al.  Multivariate Statistical Process Control with Industrial Applications , 2004, Technometrics.

[12]  H. Hotelling.  Relations Between Two Sets of Variates , 1936 .

[13]  Jay H. Lee,et al.  Partial least squares (PLS) based monitoring and control of batch digesters , 2000 .

[14]  Dexian Huang,et al.  Canonical variate analysis-based monitoring of process correlation structure using causal feature representation , 2015 .

[15]  Junwu Zhou,et al.  Robust dynamic process monitoring based on sparse representation preserving embedding , 2016 .

[16]  Jian Yang,et al.  Dynamic learning on the manifold with constrained time information and its application for dynamic process monitoring , 2017 .

[17]  C. Alippi,et al.  Simple approximation of sigmoidal functions: realistic design of digital neural networks capable of learning , 1991, 1991., IEEE International Sympoisum on Circuits and Systems.

[18]  Chih-Hung Chang,et al.  Deep and Shallow Architecture of Multilayer Neural Networks , 2015, IEEE Transactions on Neural Networks and Learning Systems.

[19]  Stéphane Mallat,et al.  A Theory for Multiresolution Signal Decomposition: The Wavelet Representation , 1989, IEEE Trans. Pattern Anal. Mach. Intell..

[20]  Soumith Chintala,et al.  Unsupervised Representation Learning with Deep Convolutional Generative Adversarial Networks , 2015, ICLR.

[21]  Jeff A. Bilmes,et al.  Deep Canonical Correlation Analysis , 2013, ICML.

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

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

[24]  Rajat Raina,et al.  Self-taught learning: transfer learning from unlabeled data , 2007, ICML '07.

[25]  Michael Möller,et al.  Learning Proximal Operators: Using Denoising Networks for Regularizing Inverse Imaging Problems , 2017, 2017 IEEE International Conference on Computer Vision (ICCV).

[26]  Tong Heng Lee,et al.  Geometrical interpretation and architecture selection of MLP , 2005, IEEE Transactions on Neural Networks.

[27]  M. J. Fuente,et al.  Fault detection and isolation in transient states using principal component analysis , 2012 .

[28]  Jie Chen,et al.  Robust Model-Based Fault Diagnosis for Dynamic Systems , 1998, The International Series on Asian Studies in Computer and Information Science.

[29]  Hongbo Shi,et al.  Time–space locality preserving coordination for multimode process monitoring , 2016 .

[30]  Yang Wang,et al.  Data-Driven Optimized Distributed Dynamic PCA for Efficient Monitoring of Large-Scale Dynamic Processes , 2017, IEEE Access.

[31]  Chenglin Wen,et al.  Fault diagnosis based on deep learning , 2016, 2016 American Control Conference (ACC).

[32]  Donghua Zhou,et al.  Recursive transformed component statistical analysis for incipient fault detection , 2017, Autom..

[33]  Shuicheng Yan,et al.  Neighborhood preserving embedding , 2005, Tenth IEEE International Conference on Computer Vision (ICCV'05) Volume 1.

[34]  Erkki Oja,et al.  Independent component analysis: algorithms and applications , 2000, Neural Networks.

[35]  Ma Yao,et al.  On-line monitoring of batch processes using generalized additive kernel principal component analysis , 2015 .

[36]  Juhan Nam,et al.  Multimodal Deep Learning , 2011, ICML.

[37]  Biao Huang,et al.  Dynamic higher-order cumulants analysis for state monitoring based on a novel lag selection , 2016, Inf. Sci..

[38]  N. Ricker.  Optimal steady-state operation of the Tennessee Eastman challenge process , 1995 .

[39]  William W. Hsieh,et al.  Nonlinear canonical correlation analysis by neural networks , 2000, Neural Networks.

[40]  Hongbo Shi,et al.  Multimode process monitoring using improved dynamic neighborhood preserving embedding , 2014 .

[41]  Bryan A. Tolson,et al.  A New Formulation for Feedforward Neural Networks , 2011, IEEE Transactions on Neural Networks.