A Denoising Based Autoassociative Model for Robust Sensor Monitoring in Nuclear Power Plants
Sensors health monitoring is essentially important for reliable functioning of safety-critical chemical and nuclear power plants. Autoassociative neural network (AANN) based empirical sensor models have widely been reported for sensor calibration monitoring. However, such ill-posed data driven model...
Saved in:
| Main Authors: | , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
Wiley
2016-01-01
|
| Series: | Science and Technology of Nuclear Installations |
| Online Access: | http://dx.doi.org/10.1155/2016/9746948 |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1832562720320258048 |
|---|---|
| author | Ahmad Shaheryar Xu-Cheng Yin Hong-Wei Hao Hazrat Ali Khalid Iqbal |
| author_facet | Ahmad Shaheryar Xu-Cheng Yin Hong-Wei Hao Hazrat Ali Khalid Iqbal |
| author_sort | Ahmad Shaheryar |
| collection | DOAJ |
| description | Sensors health monitoring is essentially important for reliable functioning of safety-critical chemical and nuclear power plants. Autoassociative neural network (AANN) based empirical sensor models have widely been reported for sensor calibration monitoring. However, such ill-posed data driven models may result in poor generalization and robustness. To address above-mentioned issues, several regularization heuristics such as training with jitter, weight decay, and cross-validation are suggested in literature. Apart from these regularization heuristics, traditional error gradient based supervised learning algorithms for multilayered AANN models are highly susceptible of being trapped in local optimum. In order to address poor regularization and robust learning issues, here, we propose a denoised autoassociative sensor model (DAASM) based on deep learning framework. Proposed DAASM model comprises multiple hidden layers which are pretrained greedily in an unsupervised fashion under denoising autoencoder architecture. In order to improve robustness, dropout heuristic and domain specific data corruption processes are exercised during unsupervised pretraining phase. The proposed sensor model is trained and tested on sensor data from a PWR type nuclear power plant. Accuracy, autosensitivity, spillover, and sequential probability ratio test (SPRT) based fault detectability metrics are used for performance assessment and comparison with extensively reported five-layer AANN model by Kramer. |
| format | Article |
| id | doaj-art-1b7292a40afd4ed59e9576a677979a47 |
| institution | Kabale University |
| issn | 1687-6075 1687-6083 |
| language | English |
| publishDate | 2016-01-01 |
| publisher | Wiley |
| record_format | Article |
| series | Science and Technology of Nuclear Installations |
| spelling | doaj-art-1b7292a40afd4ed59e9576a677979a472025-02-03T01:21:58ZengWileyScience and Technology of Nuclear Installations1687-60751687-60832016-01-01201610.1155/2016/97469489746948A Denoising Based Autoassociative Model for Robust Sensor Monitoring in Nuclear Power PlantsAhmad Shaheryar0Xu-Cheng Yin1Hong-Wei Hao2Hazrat Ali3Khalid Iqbal4University of Science and Technology Beijing (USTB), 30 Xueyuan Road, Haidian District, Beijing 100083, ChinaUniversity of Science and Technology Beijing (USTB), 30 Xueyuan Road, Haidian District, Beijing 100083, ChinaUniversity of Science and Technology Beijing (USTB), 30 Xueyuan Road, Haidian District, Beijing 100083, ChinaDepartment of Electrical Engineering, COMSATS Institute of Information Technology, University Road, Abbottabad 22060, PakistanCOMSATS Institute of Information Technology near Officers Colony, Kamra Road, Attock 43600, PakistanSensors health monitoring is essentially important for reliable functioning of safety-critical chemical and nuclear power plants. Autoassociative neural network (AANN) based empirical sensor models have widely been reported for sensor calibration monitoring. However, such ill-posed data driven models may result in poor generalization and robustness. To address above-mentioned issues, several regularization heuristics such as training with jitter, weight decay, and cross-validation are suggested in literature. Apart from these regularization heuristics, traditional error gradient based supervised learning algorithms for multilayered AANN models are highly susceptible of being trapped in local optimum. In order to address poor regularization and robust learning issues, here, we propose a denoised autoassociative sensor model (DAASM) based on deep learning framework. Proposed DAASM model comprises multiple hidden layers which are pretrained greedily in an unsupervised fashion under denoising autoencoder architecture. In order to improve robustness, dropout heuristic and domain specific data corruption processes are exercised during unsupervised pretraining phase. The proposed sensor model is trained and tested on sensor data from a PWR type nuclear power plant. Accuracy, autosensitivity, spillover, and sequential probability ratio test (SPRT) based fault detectability metrics are used for performance assessment and comparison with extensively reported five-layer AANN model by Kramer.http://dx.doi.org/10.1155/2016/9746948 |
| spellingShingle | Ahmad Shaheryar Xu-Cheng Yin Hong-Wei Hao Hazrat Ali Khalid Iqbal A Denoising Based Autoassociative Model for Robust Sensor Monitoring in Nuclear Power Plants Science and Technology of Nuclear Installations |
| title | A Denoising Based Autoassociative Model for Robust Sensor Monitoring in Nuclear Power Plants |
| title_full | A Denoising Based Autoassociative Model for Robust Sensor Monitoring in Nuclear Power Plants |
| title_fullStr | A Denoising Based Autoassociative Model for Robust Sensor Monitoring in Nuclear Power Plants |
| title_full_unstemmed | A Denoising Based Autoassociative Model for Robust Sensor Monitoring in Nuclear Power Plants |
| title_short | A Denoising Based Autoassociative Model for Robust Sensor Monitoring in Nuclear Power Plants |
| title_sort | denoising based autoassociative model for robust sensor monitoring in nuclear power plants |
| url | http://dx.doi.org/10.1155/2016/9746948 |
| work_keys_str_mv | AT ahmadshaheryar adenoisingbasedautoassociativemodelforrobustsensormonitoringinnuclearpowerplants AT xuchengyin adenoisingbasedautoassociativemodelforrobustsensormonitoringinnuclearpowerplants AT hongweihao adenoisingbasedautoassociativemodelforrobustsensormonitoringinnuclearpowerplants AT hazratali adenoisingbasedautoassociativemodelforrobustsensormonitoringinnuclearpowerplants AT khalidiqbal adenoisingbasedautoassociativemodelforrobustsensormonitoringinnuclearpowerplants AT ahmadshaheryar denoisingbasedautoassociativemodelforrobustsensormonitoringinnuclearpowerplants AT xuchengyin denoisingbasedautoassociativemodelforrobustsensormonitoringinnuclearpowerplants AT hongweihao denoisingbasedautoassociativemodelforrobustsensormonitoringinnuclearpowerplants AT hazratali denoisingbasedautoassociativemodelforrobustsensormonitoringinnuclearpowerplants AT khalidiqbal denoisingbasedautoassociativemodelforrobustsensormonitoringinnuclearpowerplants |