Fault Detection and Diagnosis in Industrial Processes with Variational Autoencoder: A Comprehensive Study
This work considers industrial process monitoring using a variational autoencoder (VAE). As a powerful deep generative model, the variational autoencoder and its variants have become popular for process monitoring. However, its monitoring ability, especially its fault diagnosis ability, has not been...
Main Authors: | , , |
---|---|
Format: | Article |
Language: | English |
Published: |
MDPI AG
2021-12-01
|
Series: | Sensors |
Subjects: | |
Online Access: | https://www.mdpi.com/1424-8220/22/1/227 |
_version_ | 1797497609316728832 |
---|---|
author | Jinlin Zhu Muyun Jiang Zhong Liu |
author_facet | Jinlin Zhu Muyun Jiang Zhong Liu |
author_sort | Jinlin Zhu |
collection | DOAJ |
description | This work considers industrial process monitoring using a variational autoencoder (VAE). As a powerful deep generative model, the variational autoencoder and its variants have become popular for process monitoring. However, its monitoring ability, especially its fault diagnosis ability, has not been well investigated. In this paper, the process modeling and monitoring capabilities of several VAE variants are comprehensively studied. First, fault detection schemes are defined in three distinct ways, considering latent, residual, and the combined domains. Afterwards, to conduct the fault diagnosis, we first define the deep contribution plot, and then a deep reconstruction-based contribution diagram is proposed for deep domains under the fault propagation mechanism. In a case study, the performance of the process monitoring capability of four deep VAE models, namely, the static VAE model, the dynamic VAE model, and the recurrent VAE models (LSTM-VAE and GRU-VAE), has been comparatively evaluated on the industrial benchmark Tennessee Eastman process. Results show that recurrent VAEs with a deep reconstruction-based diagnosis mechanism are recommended for industrial process monitoring tasks. |
first_indexed | 2024-03-10T03:21:41Z |
format | Article |
id | doaj.art-ff1ad6c5007c456ab8d1822f6af636e2 |
institution | Directory Open Access Journal |
issn | 1424-8220 |
language | English |
last_indexed | 2024-03-10T03:21:41Z |
publishDate | 2021-12-01 |
publisher | MDPI AG |
record_format | Article |
series | Sensors |
spelling | doaj.art-ff1ad6c5007c456ab8d1822f6af636e22023-11-23T12:18:57ZengMDPI AGSensors1424-82202021-12-0122122710.3390/s22010227Fault Detection and Diagnosis in Industrial Processes with Variational Autoencoder: A Comprehensive StudyJinlin Zhu0Muyun Jiang1Zhong Liu2State Key Laboratory of Food Science and Technology, Jiangnan University, Wuxi 214122, ChinaSchool of Computer Science and Engineering, Nanyang Technological University, Singapore 639798, SingaporeKey Laboratory of Advanced Process Control for Light Industry (Ministry of Education), Jiangnan University, Wuxi 214122, ChinaThis work considers industrial process monitoring using a variational autoencoder (VAE). As a powerful deep generative model, the variational autoencoder and its variants have become popular for process monitoring. However, its monitoring ability, especially its fault diagnosis ability, has not been well investigated. In this paper, the process modeling and monitoring capabilities of several VAE variants are comprehensively studied. First, fault detection schemes are defined in three distinct ways, considering latent, residual, and the combined domains. Afterwards, to conduct the fault diagnosis, we first define the deep contribution plot, and then a deep reconstruction-based contribution diagram is proposed for deep domains under the fault propagation mechanism. In a case study, the performance of the process monitoring capability of four deep VAE models, namely, the static VAE model, the dynamic VAE model, and the recurrent VAE models (LSTM-VAE and GRU-VAE), has been comparatively evaluated on the industrial benchmark Tennessee Eastman process. Results show that recurrent VAEs with a deep reconstruction-based diagnosis mechanism are recommended for industrial process monitoring tasks.https://www.mdpi.com/1424-8220/22/1/227process monitoringdeep modelvariational autoencoderdeep reconstructiondynamic process |
spellingShingle | Jinlin Zhu Muyun Jiang Zhong Liu Fault Detection and Diagnosis in Industrial Processes with Variational Autoencoder: A Comprehensive Study Sensors process monitoring deep model variational autoencoder deep reconstruction dynamic process |
title | Fault Detection and Diagnosis in Industrial Processes with Variational Autoencoder: A Comprehensive Study |
title_full | Fault Detection and Diagnosis in Industrial Processes with Variational Autoencoder: A Comprehensive Study |
title_fullStr | Fault Detection and Diagnosis in Industrial Processes with Variational Autoencoder: A Comprehensive Study |
title_full_unstemmed | Fault Detection and Diagnosis in Industrial Processes with Variational Autoencoder: A Comprehensive Study |
title_short | Fault Detection and Diagnosis in Industrial Processes with Variational Autoencoder: A Comprehensive Study |
title_sort | fault detection and diagnosis in industrial processes with variational autoencoder a comprehensive study |
topic | process monitoring deep model variational autoencoder deep reconstruction dynamic process |
url | https://www.mdpi.com/1424-8220/22/1/227 |
work_keys_str_mv | AT jinlinzhu faultdetectionanddiagnosisinindustrialprocesseswithvariationalautoencoderacomprehensivestudy AT muyunjiang faultdetectionanddiagnosisinindustrialprocesseswithvariationalautoencoderacomprehensivestudy AT zhongliu faultdetectionanddiagnosisinindustrialprocesseswithvariationalautoencoderacomprehensivestudy |