Time Series Recovery Using Adjacent Channel Data Based on LSTM: A Case Study of Subway Vibrations
Multi-sensor technology has been widely applied in the condition monitoring of rail transit. In practice, the data of some channels in the high channel counts are often abnormal or lost due to the abnormality and damage of the sensors, thus resulting in a large amount of data waste. A method for the...
| Main Authors: | , , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
MDPI AG
2022-11-01
|
| Series: | Applied Sciences |
| Subjects: | |
| Online Access: | https://www.mdpi.com/2076-3417/12/22/11497 |
| _version_ | 1797466004091043840 |
|---|---|
| author | Tao Xin Yi Yang Xiaoli Zheng Jing Lin Sen Wang Pengsong Wang |
| author_facet | Tao Xin Yi Yang Xiaoli Zheng Jing Lin Sen Wang Pengsong Wang |
| author_sort | Tao Xin |
| collection | DOAJ |
| description | Multi-sensor technology has been widely applied in the condition monitoring of rail transit. In practice, the data of some channels in the high channel counts are often abnormal or lost due to the abnormality and damage of the sensors, thus resulting in a large amount of data waste. A method for the data recovery of lost channels by using adjacent channel data is proposed to solve this problem. Based on the LSTM network algorithm, a data recovery model is established based on the “sequence-to-sequence” regression analysis of adjacent channel data. Taking the measured vibration data of a subway as an example, the network is trained with multi-channel measured data to recover the lost channel data of time-series characteristics. The results show that this multi-channel data recovery model is feasible, and the accuracy is up to 98%. This method can also further reduce the number of channels that need to be collected. |
| first_indexed | 2024-03-09T18:30:37Z |
| format | Article |
| id | doaj.art-bc8efab794a94e028e1d513892088a63 |
| institution | Directory Open Access Journal |
| issn | 2076-3417 |
| language | English |
| last_indexed | 2024-03-09T18:30:37Z |
| publishDate | 2022-11-01 |
| publisher | MDPI AG |
| record_format | Article |
| series | Applied Sciences |
| spelling | doaj.art-bc8efab794a94e028e1d513892088a632023-11-24T07:36:16ZengMDPI AGApplied Sciences2076-34172022-11-0112221149710.3390/app122211497Time Series Recovery Using Adjacent Channel Data Based on LSTM: A Case Study of Subway VibrationsTao Xin0Yi Yang1Xiaoli Zheng2Jing Lin3Sen Wang4Pengsong Wang5School of Civil Engineering, Beijing Jiaotong University, Beijing 100044, ChinaSchool of Civil Engineering, Beijing Jiaotong University, Beijing 100044, ChinaBeijing Municipal Institute of City Planning and Design, Beijing 100045, ChinaDepartment of Civil, Environmental and Natural Resources Engineering, Luleå University of Technology, 97187 Luleå, SwedenSchool of Civil Engineering, Beijing Jiaotong University, Beijing 100044, ChinaSchool of Civil Engineering, Beijing Jiaotong University, Beijing 100044, ChinaMulti-sensor technology has been widely applied in the condition monitoring of rail transit. In practice, the data of some channels in the high channel counts are often abnormal or lost due to the abnormality and damage of the sensors, thus resulting in a large amount of data waste. A method for the data recovery of lost channels by using adjacent channel data is proposed to solve this problem. Based on the LSTM network algorithm, a data recovery model is established based on the “sequence-to-sequence” regression analysis of adjacent channel data. Taking the measured vibration data of a subway as an example, the network is trained with multi-channel measured data to recover the lost channel data of time-series characteristics. The results show that this multi-channel data recovery model is feasible, and the accuracy is up to 98%. This method can also further reduce the number of channels that need to be collected.https://www.mdpi.com/2076-3417/12/22/11497multi-channel datatime-series recoveryneural networkregression analysisdata recoverytime domain |
| spellingShingle | Tao Xin Yi Yang Xiaoli Zheng Jing Lin Sen Wang Pengsong Wang Time Series Recovery Using Adjacent Channel Data Based on LSTM: A Case Study of Subway Vibrations Applied Sciences multi-channel data time-series recovery neural network regression analysis data recovery time domain |
| title | Time Series Recovery Using Adjacent Channel Data Based on LSTM: A Case Study of Subway Vibrations |
| title_full | Time Series Recovery Using Adjacent Channel Data Based on LSTM: A Case Study of Subway Vibrations |
| title_fullStr | Time Series Recovery Using Adjacent Channel Data Based on LSTM: A Case Study of Subway Vibrations |
| title_full_unstemmed | Time Series Recovery Using Adjacent Channel Data Based on LSTM: A Case Study of Subway Vibrations |
| title_short | Time Series Recovery Using Adjacent Channel Data Based on LSTM: A Case Study of Subway Vibrations |
| title_sort | time series recovery using adjacent channel data based on lstm a case study of subway vibrations |
| topic | multi-channel data time-series recovery neural network regression analysis data recovery time domain |
| url | https://www.mdpi.com/2076-3417/12/22/11497 |
| work_keys_str_mv | AT taoxin timeseriesrecoveryusingadjacentchanneldatabasedonlstmacasestudyofsubwayvibrations AT yiyang timeseriesrecoveryusingadjacentchanneldatabasedonlstmacasestudyofsubwayvibrations AT xiaolizheng timeseriesrecoveryusingadjacentchanneldatabasedonlstmacasestudyofsubwayvibrations AT jinglin timeseriesrecoveryusingadjacentchanneldatabasedonlstmacasestudyofsubwayvibrations AT senwang timeseriesrecoveryusingadjacentchanneldatabasedonlstmacasestudyofsubwayvibrations AT pengsongwang timeseriesrecoveryusingadjacentchanneldatabasedonlstmacasestudyofsubwayvibrations |