Deep learning based nonlinear principal component analysis for industrial process fault detection
Deep learning based nonlinear principal component analysis for industrial process fault detection
Principal component analysis (PCA) and kernel PCA (KPCA) are the state-of-art machine learning methods widely used in industrial process monitoring and fault detection field. However, these methods build shallow statistical models based on single layer of features and may not achieve the best monitoring performance. In order to sufficiently mine the intrinsic data features, a deep learning based nonlinear PCA method, referred to as deep PCA (DePCA), is proposed in this paper. Motivated by the idea of deep learning, a layer-wise statistical model structure is designed to extract multi-layer data features, including both linear and nonlinear principal components. At each layer, two monitoring statistics are constructed to monitor the feature changes. For integrating the monitoring statistics of all feature layers, a Bayesian inference strategy is applied to convert the monitoring statistics into fault probabilities, which are weighted to form two probability-based comprehensive monitoring statistics for process fault detection. A case study using the benchmark Tennessee Eastman process demonstrates the superior performance of the proposed DePCA method over the traditional PCA and KPCA methods.
1-7
Deng, Xiaogang
537ff6f8-8fb1-4eb2-b5e9-c0eec6ef0093
Tian, Xuemin
982a5a50-8672-4772-b877-7dc9bc64264b
Chen, Sheng
9310a111-f79a-48b8-98c7-383ca93cbb80
Harris, Chris
693153ea-2370-4d4c-ab05-c85ef551607a
14 May 2017
Deng, Xiaogang
537ff6f8-8fb1-4eb2-b5e9-c0eec6ef0093
Tian, Xuemin
982a5a50-8672-4772-b877-7dc9bc64264b
Chen, Sheng
9310a111-f79a-48b8-98c7-383ca93cbb80
Harris, Chris
693153ea-2370-4d4c-ab05-c85ef551607a
Deng, Xiaogang, Tian, Xuemin, Chen, Sheng and Harris, Chris
(2017)
Deep learning based nonlinear principal component analysis for industrial process fault detection.
In 2017 International Joint Conference on Neural Networks.
IEEE.
.
(doi:10.1109/IJCNN.2017.7965994).
Record type:
Conference or Workshop Item
(Paper)
Abstract
Principal component analysis (PCA) and kernel PCA (KPCA) are the state-of-art machine learning methods widely used in industrial process monitoring and fault detection field. However, these methods build shallow statistical models based on single layer of features and may not achieve the best monitoring performance. In order to sufficiently mine the intrinsic data features, a deep learning based nonlinear PCA method, referred to as deep PCA (DePCA), is proposed in this paper. Motivated by the idea of deep learning, a layer-wise statistical model structure is designed to extract multi-layer data features, including both linear and nonlinear principal components. At each layer, two monitoring statistics are constructed to monitor the feature changes. For integrating the monitoring statistics of all feature layers, a Bayesian inference strategy is applied to convert the monitoring statistics into fault probabilities, which are weighted to form two probability-based comprehensive monitoring statistics for process fault detection. A case study using the benchmark Tennessee Eastman process demonstrates the superior performance of the proposed DePCA method over the traditional PCA and KPCA methods.
More information
Published date: 14 May 2017
Venue - Dates:
International Joint Conference on Neural Networks (IJCNN 2017), , Anchorage, United States, 2017-05-14 - 2017-05-19
Organisations:
Southampton Wireless Group
Identifiers
Local EPrints ID: 411641
URI: http://eprints.soton.ac.uk/id/eprint/411641
PURE UUID: 549c9ec5-7cd1-4174-abe5-aaa981b75b90
Catalogue record
Date deposited: 21 Jun 2017 16:32
Last modified: 15 Mar 2024 13:48
Export record
Altmetrics
Contributors
Author:
Xiaogang Deng
Author:
Xuemin Tian
Author:
Sheng Chen
Author:
Chris Harris
Download statistics
Downloads from ePrints over the past year. Other digital versions may also be available to download e.g. from the publisher's website.
View more statistics