Visualization analysis for fault diagnosis in chemical processes using recurrent neural networks

Abstract The mechanism of classification of the RNNs was revealed in this study. The benchmark of the Tennessee Eastman process was used to demonstrate the performance of the RNN-based fault diagnosis model. This study investigated the fault diagnosis performance of the entire and early detections for complicated chemical processes using the time-series recurrent neural networks (RNNs). The investigation included various layers and neuron nodes in RNNs using lean and rich training datasets and compared these RNNs with the artificial neural networks (ANNs). The results showed that the RNNs had better classification accuracies than the ANNs regardless of using lean or rich training datasets. The general classification mechanism was a priori classification that used normal operating data as the center so that it was incapable of separating the fault types having similar features of the normal operating data. RNNs drove the normal operating data out of the center and created the even spatial distribution of fault types, leading that RNNs were effective in classifying the fault types with subtle features when there is sufficient data.

[1]  Yoshua Bengio,et al.  Deep Sparse Rectifier Neural Networks , 2011, AISTATS.

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

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

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

[5]  Lovekesh Vig,et al.  Long Short Term Memory Networks for Anomaly Detection in Time Series , 2015, ESANN.

[6]  Xiaogang Deng,et al.  Multimode non‐Gaussian process monitoring based on local entropy independent component analysis , 2017 .

[7]  Hao Wu,et al.  Deep convolutional neural network model based chemical process fault diagnosis , 2018, Comput. Chem. Eng..

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

[9]  Venkat Venkatasubramanian,et al.  PCA-SDG based process monitoring and fault diagnosis , 1999 .

[10]  Xiaoqiang Lu,et al.  Scene Recognition by Manifold Regularized Deep Learning Architecture , 2015, IEEE Transactions on Neural Networks and Learning Systems.

[11]  Seongkyu Yoon,et al.  Principal‐component analysis of multiscale data for process monitoring and fault diagnosis , 2004 .

[12]  Nola D. Tracy,et al.  Multivariate Control Charts for Individual Observations , 1992 .

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

[14]  Dexian Huang,et al.  Slow feature analysis for monitoring and diagnosis of control performance , 2016 .

[15]  Jinsong Zhao,et al.  A new unsupervised data mining method based on the stacked autoencoder for chemical process fault diagnosis , 2020, Comput. Chem. Eng..

[16]  Jinsong Zhao,et al.  A novel process monitoring approach based on variational recurrent autoencoder , 2019, Comput. Chem. Eng..

[17]  Geoffrey E. Hinton,et al.  Visualizing Data using t-SNE , 2008 .

[18]  Chen Jing,et al.  SVM and PCA based fault classification approaches for complicated industrial process , 2015, Neurocomputing.

[19]  S. Joe Qin,et al.  Multivariate process monitoring and fault diagnosis by multi-scale PCA , 2002 .

[20]  Raghunathan Rengaswamy,et al.  Fault diagnosis using dynamic trend analysis: A review and recent developments , 2007, Eng. Appl. Artif. Intell..

[21]  Zhanpeng Zhang,et al.  A deep belief network based fault diagnosis model for complex chemical processes , 2017, Comput. Chem. Eng..

[22]  John F. MacGregor,et al.  Process monitoring and diagnosis by multiblock PLS methods , 1994 .