Anomaly Detection of Power Plant Equipment Using Long Short-Term Memory Based Autoencoder Neural Network
Anomaly detection is of great significance in condition-based maintenance of power plant equipment. The conventional fixed threshold detection method is not able to perform early detection of equipment abnormalities. In this study, a general anomaly detection framework based on a long short-term mem...
| Main Authors: | , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
MDPI AG
2020-10-01
|
| Series: | Sensors |
| Subjects: | |
| Online Access: | https://www.mdpi.com/1424-8220/20/21/6164 |
| _version_ | 1797549476998545408 |
|---|---|
| author | Di Hu Chen Zhang Tao Yang Gang Chen |
| author_facet | Di Hu Chen Zhang Tao Yang Gang Chen |
| author_sort | Di Hu |
| collection | DOAJ |
| description | Anomaly detection is of great significance in condition-based maintenance of power plant equipment. The conventional fixed threshold detection method is not able to perform early detection of equipment abnormalities. In this study, a general anomaly detection framework based on a long short-term memory-based autoencoder (LSTM-AE) network is proposed. A normal behavior model (NBM) is established to learn the normal behavior patterns of the operating variables of the equipment in space and time. Based on the similarity analysis between the NBM output distribution and the corresponding measurement distribution, the Mahalanobis distance (MD) is used to describe the overall residual (OR) of the model. The reasonable range is obtained using kernel density estimation (KDE) with a 99% confidence interval, and the OR is monitored to detect abnormalities in real-time. An induced draft fan is chosen as a case study. Results show that the established NBM has excellent accuracy and generalizability, with average root mean square errors of 0.026 and 0.035 for the training and test data, respectively, and average mean absolute percentage errors of 0.027%. Moreover, the abnormal operation case shows that the proposed framework can be effectively used for the early detection of abnormalities. |
| first_indexed | 2024-03-10T15:15:10Z |
| format | Article |
| id | doaj.art-c88808cd28a94889a36409beeebaf388 |
| institution | Directory Open Access Journal |
| issn | 1424-8220 |
| language | English |
| last_indexed | 2024-03-10T15:15:10Z |
| publishDate | 2020-10-01 |
| publisher | MDPI AG |
| record_format | Article |
| series | Sensors |
| spelling | doaj.art-c88808cd28a94889a36409beeebaf3882023-11-20T19:02:13ZengMDPI AGSensors1424-82202020-10-012021616410.3390/s20216164Anomaly Detection of Power Plant Equipment Using Long Short-Term Memory Based Autoencoder Neural NetworkDi Hu0Chen Zhang1Tao Yang2Gang Chen3School of Energy and Power Engineering, Huazhong University of Science and Technology, Wuhan 430074, ChinaSchool of Energy and Power Engineering, Huazhong University of Science and Technology, Wuhan 430074, ChinaSchool of Energy and Power Engineering, Huazhong University of Science and Technology, Wuhan 430074, ChinaSchool of Energy and Power Engineering, Huazhong University of Science and Technology, Wuhan 430074, ChinaAnomaly detection is of great significance in condition-based maintenance of power plant equipment. The conventional fixed threshold detection method is not able to perform early detection of equipment abnormalities. In this study, a general anomaly detection framework based on a long short-term memory-based autoencoder (LSTM-AE) network is proposed. A normal behavior model (NBM) is established to learn the normal behavior patterns of the operating variables of the equipment in space and time. Based on the similarity analysis between the NBM output distribution and the corresponding measurement distribution, the Mahalanobis distance (MD) is used to describe the overall residual (OR) of the model. The reasonable range is obtained using kernel density estimation (KDE) with a 99% confidence interval, and the OR is monitored to detect abnormalities in real-time. An induced draft fan is chosen as a case study. Results show that the established NBM has excellent accuracy and generalizability, with average root mean square errors of 0.026 and 0.035 for the training and test data, respectively, and average mean absolute percentage errors of 0.027%. Moreover, the abnormal operation case shows that the proposed framework can be effectively used for the early detection of abnormalities.https://www.mdpi.com/1424-8220/20/21/6164anomaly detectionpower plantartificial neural networkslong short-term memory based autoencoder neural networksnormal behavior model |
| spellingShingle | Di Hu Chen Zhang Tao Yang Gang Chen Anomaly Detection of Power Plant Equipment Using Long Short-Term Memory Based Autoencoder Neural Network Sensors anomaly detection power plant artificial neural networks long short-term memory based autoencoder neural networks normal behavior model |
| title | Anomaly Detection of Power Plant Equipment Using Long Short-Term Memory Based Autoencoder Neural Network |
| title_full | Anomaly Detection of Power Plant Equipment Using Long Short-Term Memory Based Autoencoder Neural Network |
| title_fullStr | Anomaly Detection of Power Plant Equipment Using Long Short-Term Memory Based Autoencoder Neural Network |
| title_full_unstemmed | Anomaly Detection of Power Plant Equipment Using Long Short-Term Memory Based Autoencoder Neural Network |
| title_short | Anomaly Detection of Power Plant Equipment Using Long Short-Term Memory Based Autoencoder Neural Network |
| title_sort | anomaly detection of power plant equipment using long short term memory based autoencoder neural network |
| topic | anomaly detection power plant artificial neural networks long short-term memory based autoencoder neural networks normal behavior model |
| url | https://www.mdpi.com/1424-8220/20/21/6164 |
| work_keys_str_mv | AT dihu anomalydetectionofpowerplantequipmentusinglongshorttermmemorybasedautoencoderneuralnetwork AT chenzhang anomalydetectionofpowerplantequipmentusinglongshorttermmemorybasedautoencoderneuralnetwork AT taoyang anomalydetectionofpowerplantequipmentusinglongshorttermmemorybasedautoencoderneuralnetwork AT gangchen anomalydetectionofpowerplantequipmentusinglongshorttermmemorybasedautoencoderneuralnetwork |