The Next Failure Time Prediction of Escalators via Deep Neural Network with Dynamic Time Warping Preprocessing
The escalator is one of the most popular travel methods in public places, and the safe working of the escalator is significant. Accurately predicting the escalator failure time can provide scientific guidance for maintenance to avoid accidents. However, failure data have features of short length, no...
| Main Authors: | , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
MDPI AG
2020-08-01
|
| Series: | Applied Sciences |
| Subjects: | |
| Online Access: | https://www.mdpi.com/2076-3417/10/16/5622 |
| id |
doaj-c3676a2d84d14a7d92eb38ab8bd19418 |
|---|---|
| record_format |
Article |
| spelling |
doaj-c3676a2d84d14a7d92eb38ab8bd194182020-11-25T03:37:54ZengMDPI AGApplied Sciences2076-34172020-08-01105622562210.3390/app10165622The Next Failure Time Prediction of Escalators via Deep Neural Network with Dynamic Time Warping PreprocessingZitong Zhou0Yanyang Zi1Jingsong Xie2Jinglong Chen3Tong An4State Key Laboratory for Manufacturing and Systems Engineering, Xi’an Jiaotong University, Xi’an 710049, ChinaState Key Laboratory for Manufacturing and Systems Engineering, Xi’an Jiaotong University, Xi’an 710049, ChinaSchool of Traffic and Transportation Engineering, Central South University, Changsha 410075, ChinaState Key Laboratory for Manufacturing and Systems Engineering, Xi’an Jiaotong University, Xi’an 710049, ChinaState Key Laboratory for Manufacturing and Systems Engineering, Xi’an Jiaotong University, Xi’an 710049, ChinaThe escalator is one of the most popular travel methods in public places, and the safe working of the escalator is significant. Accurately predicting the escalator failure time can provide scientific guidance for maintenance to avoid accidents. However, failure data have features of short length, non-uniform sampling, and random interference, which makes the data modeling difficult. Therefore, a strategy that combines data quality enhancement with deep neural networks is proposed for escalator failure time prediction in this paper. First, a comprehensive selection indicator (CSI) that can describe the stationarity and complexity of time series is established to select inherently excellent failure sequences. According to the CSI, failure sequences with high stationarity and low complexity are selected as the referenced sequences to enhance the quality of other failure sequences by using dynamic time warping preprocessing. Secondly, a deep neural network combining the advantages of a convolutional neural network and long short-term memory is built to train and predict quality-enhanced failure sequences. Finally, the failure-recall record of six escalators used for 6 years is analyzed by using the proposed method as a case study, and the results show that the proposed method can reduce the average prediction error of failure time to less than one month.https://www.mdpi.com/2076-3417/10/16/5622failure time predictionconvolutional neural networklong-short term memorydynamic time warpingescalator |
| collection |
DOAJ |
| language |
English |
| format |
Article |
| sources |
DOAJ |
| author |
Zitong Zhou Yanyang Zi Jingsong Xie Jinglong Chen Tong An |
| spellingShingle |
Zitong Zhou Yanyang Zi Jingsong Xie Jinglong Chen Tong An The Next Failure Time Prediction of Escalators via Deep Neural Network with Dynamic Time Warping Preprocessing Applied Sciences failure time prediction convolutional neural network long-short term memory dynamic time warping escalator |
| author_facet |
Zitong Zhou Yanyang Zi Jingsong Xie Jinglong Chen Tong An |
| author_sort |
Zitong Zhou |
| title |
The Next Failure Time Prediction of Escalators via Deep Neural Network with Dynamic Time Warping Preprocessing |
| title_short |
The Next Failure Time Prediction of Escalators via Deep Neural Network with Dynamic Time Warping Preprocessing |
| title_full |
The Next Failure Time Prediction of Escalators via Deep Neural Network with Dynamic Time Warping Preprocessing |
| title_fullStr |
The Next Failure Time Prediction of Escalators via Deep Neural Network with Dynamic Time Warping Preprocessing |
| title_full_unstemmed |
The Next Failure Time Prediction of Escalators via Deep Neural Network with Dynamic Time Warping Preprocessing |
| title_sort |
next failure time prediction of escalators via deep neural network with dynamic time warping preprocessing |
| publisher |
MDPI AG |
| series |
Applied Sciences |
| issn |
2076-3417 |
| publishDate |
2020-08-01 |
| description |
The escalator is one of the most popular travel methods in public places, and the safe working of the escalator is significant. Accurately predicting the escalator failure time can provide scientific guidance for maintenance to avoid accidents. However, failure data have features of short length, non-uniform sampling, and random interference, which makes the data modeling difficult. Therefore, a strategy that combines data quality enhancement with deep neural networks is proposed for escalator failure time prediction in this paper. First, a comprehensive selection indicator (CSI) that can describe the stationarity and complexity of time series is established to select inherently excellent failure sequences. According to the CSI, failure sequences with high stationarity and low complexity are selected as the referenced sequences to enhance the quality of other failure sequences by using dynamic time warping preprocessing. Secondly, a deep neural network combining the advantages of a convolutional neural network and long short-term memory is built to train and predict quality-enhanced failure sequences. Finally, the failure-recall record of six escalators used for 6 years is analyzed by using the proposed method as a case study, and the results show that the proposed method can reduce the average prediction error of failure time to less than one month. |
| topic |
failure time prediction convolutional neural network long-short term memory dynamic time warping escalator |
| url |
https://www.mdpi.com/2076-3417/10/16/5622 |
| work_keys_str_mv |
AT zitongzhou thenextfailuretimepredictionofescalatorsviadeepneuralnetworkwithdynamictimewarpingpreprocessing AT yanyangzi thenextfailuretimepredictionofescalatorsviadeepneuralnetworkwithdynamictimewarpingpreprocessing AT jingsongxie thenextfailuretimepredictionofescalatorsviadeepneuralnetworkwithdynamictimewarpingpreprocessing AT jinglongchen thenextfailuretimepredictionofescalatorsviadeepneuralnetworkwithdynamictimewarpingpreprocessing AT tongan thenextfailuretimepredictionofescalatorsviadeepneuralnetworkwithdynamictimewarpingpreprocessing AT zitongzhou nextfailuretimepredictionofescalatorsviadeepneuralnetworkwithdynamictimewarpingpreprocessing AT yanyangzi nextfailuretimepredictionofescalatorsviadeepneuralnetworkwithdynamictimewarpingpreprocessing AT jingsongxie nextfailuretimepredictionofescalatorsviadeepneuralnetworkwithdynamictimewarpingpreprocessing AT jinglongchen nextfailuretimepredictionofescalatorsviadeepneuralnetworkwithdynamictimewarpingpreprocessing AT tongan nextfailuretimepredictionofescalatorsviadeepneuralnetworkwithdynamictimewarpingpreprocessing |
| _version_ |
1724543113497673728 |