Unsupervised Transfer Learning for Remaining Useful Life Prediction of Elevator Brake

In order to improve the life prediction effect of elevator brake in the real working environment, an unsupervised deep transfer learning (UDTL) method based on long short-term memory encoder-decoder (LSTM-ED) was proposed. The simulation data were used to analyze the health status of brake when it w...

Descripción completa

Guardado en:

Detalles Bibliográficos
Autor principal:	JIANG Yudi, HU Hui, YIN Yuehong
Formato:	article
Lenguaje:	ZH
Publicado:	Editorial Office of Journal of Shanghai Jiao Tong University 2021
Materias:	elevator brake unsupervised deep transfer learning (udtl) long short-term memory encoder-decoder (lstm-ed) remaining useful life (rul) step training Engineering (General). Civil engineering (General) TA1-2040 Chemical engineering TP155-156 Naval architecture. Shipbuilding. Marine engineering VM1-989
Acceso en línea:	https://doaj.org/article/a2a1a5e4197e4338aaa2e83177871b30
Etiquetas:	Agregar Etiqueta Sin Etiquetas, Sea el primero en etiquetar este registro!

id	oai:doaj.org-article:a2a1a5e4197e4338aaa2e83177871b30
record_format	dspace
spelling	oai:doaj.org-article:a2a1a5e4197e4338aaa2e83177871b302021-12-03T02:59:23ZUnsupervised Transfer Learning for Remaining Useful Life Prediction of Elevator Brake1006-246710.16183/j.cnki.jsjtu.2020.175https://doaj.org/article/a2a1a5e4197e4338aaa2e83177871b302021-11-01T00:00:00Zhttp://xuebao.sjtu.edu.cn/article/2021/1006-2467/1006-2467-55-11-1408.shtmlhttps://doaj.org/toc/1006-2467In order to improve the life prediction effect of elevator brake in the real working environment, an unsupervised deep transfer learning (UDTL) method based on long short-term memory encoder-decoder (LSTM-ED) was proposed. The simulation data were used to analyze the health status of brake when it was working. First, the LSTM-ED and the fully connected network were initially trained through the source domain data. Then, the LSTM-ED was used as a feature extractor to map the simulated and actual data to the feature space, and the maximum mean discrepancy was adopted to achieve data alignment. Finally, the target domain data in the feature space was regressed through the fully connected network to predict the remaining useful life (RUL) of the real brake. In the training phase, a step-by-step training method was used to ensure the accuracy of a single module. The validity was verified by comparing the experimental simulation data with the real working data in the elevator tower. The results show that by introducing the transfer learning and step-by-step training methods, the proposed method can reduce the mean square error of RUL prediction to 0.0016, and can achieve accurate RUL prediction of elevator brakes in real working environment.JIANG Yudi, HU Hui, YIN YuehongEditorial Office of Journal of Shanghai Jiao Tong Universityarticleelevator brakeunsupervised deep transfer learning (udtl)long short-term memory encoder-decoder (lstm-ed)remaining useful life (rul)step trainingEngineering (General). Civil engineering (General)TA1-2040Chemical engineeringTP155-156Naval architecture. Shipbuilding. Marine engineeringVM1-989ZHShanghai Jiaotong Daxue xuebao, Vol 55, Iss 11, Pp 1408-1416 (2021)
institution	DOAJ
collection	DOAJ
language	ZH
topic	elevator brake unsupervised deep transfer learning (udtl) long short-term memory encoder-decoder (lstm-ed) remaining useful life (rul) step training Engineering (General). Civil engineering (General) TA1-2040 Chemical engineering TP155-156 Naval architecture. Shipbuilding. Marine engineering VM1-989
spellingShingle	elevator brake unsupervised deep transfer learning (udtl) long short-term memory encoder-decoder (lstm-ed) remaining useful life (rul) step training Engineering (General). Civil engineering (General) TA1-2040 Chemical engineering TP155-156 Naval architecture. Shipbuilding. Marine engineering VM1-989 JIANG Yudi, HU Hui, YIN Yuehong Unsupervised Transfer Learning for Remaining Useful Life Prediction of Elevator Brake
description	In order to improve the life prediction effect of elevator brake in the real working environment, an unsupervised deep transfer learning (UDTL) method based on long short-term memory encoder-decoder (LSTM-ED) was proposed. The simulation data were used to analyze the health status of brake when it was working. First, the LSTM-ED and the fully connected network were initially trained through the source domain data. Then, the LSTM-ED was used as a feature extractor to map the simulated and actual data to the feature space, and the maximum mean discrepancy was adopted to achieve data alignment. Finally, the target domain data in the feature space was regressed through the fully connected network to predict the remaining useful life (RUL) of the real brake. In the training phase, a step-by-step training method was used to ensure the accuracy of a single module. The validity was verified by comparing the experimental simulation data with the real working data in the elevator tower. The results show that by introducing the transfer learning and step-by-step training methods, the proposed method can reduce the mean square error of RUL prediction to 0.0016, and can achieve accurate RUL prediction of elevator brakes in real working environment.
format	article
author	JIANG Yudi, HU Hui, YIN Yuehong
author_facet	JIANG Yudi, HU Hui, YIN Yuehong
author_sort	JIANG Yudi, HU Hui, YIN Yuehong
title	Unsupervised Transfer Learning for Remaining Useful Life Prediction of Elevator Brake
title_short	Unsupervised Transfer Learning for Remaining Useful Life Prediction of Elevator Brake
title_full	Unsupervised Transfer Learning for Remaining Useful Life Prediction of Elevator Brake
title_fullStr	Unsupervised Transfer Learning for Remaining Useful Life Prediction of Elevator Brake
title_full_unstemmed	Unsupervised Transfer Learning for Remaining Useful Life Prediction of Elevator Brake
title_sort	unsupervised transfer learning for remaining useful life prediction of elevator brake
publisher	Editorial Office of Journal of Shanghai Jiao Tong University
publishDate	2021
url	https://doaj.org/article/a2a1a5e4197e4338aaa2e83177871b30
work_keys_str_mv	AT jiangyudihuhuiyinyuehong unsupervisedtransferlearningforremainingusefullifepredictionofelevatorbrake
_version_	1718373948567060480

Unsupervised Transfer Learning for Remaining Useful Life Prediction of Elevator Brake

Ejemplares similares