Deep Learning for Track Quality Evaluation of High-Speed Railway Based on Vehicle-Body Vibration Prediction
Document identifier: oai:DiVA.org:ltu-77232
Access full text here:
10.1109/ACCESS.2019.2960537Keyword: Engineering and Technology,
Civil Engineering,
Other Civil Engineering,
Teknik och teknologier,
Samhällsbyggnadsteknik,
Annan samhällsbyggnadsteknik,
Track quality evaluation,
Track geometry,
Vehicle-body vibration,
Convolutional neural network (CNN),
Long short-term memory (LSTM),
CNN-LSTM,
Drift och underhållsteknik,
Operation and MaintenancePublication year: 2020Relevant Sustainable Development Goals (SDGs):
The SDG label(s) above have been assigned by OSDG.aiAbstract: Track quality evaluation is fundamental for track maintenance. Around the world, track geometry standards are established to evaluate track quality. However, these standards may not be capable of detecting some abnormal track geometry conditions that can cause considerable vehicle-body vibration. And people gradually realized that track quality evaluation should be based not only on track geometry but also on vehicle performance. Vehicle-body vibration prediction is beneficial for locating potential track geometry defects, and the predicted accelerations can be used as an auxiliary index for assessing track quality. For this purpose, this paper gives a method to predict vehicle-body vibration based on deep learning, which represents one of the newest areas in artificial intelligence. By integrating convolutional neural network (CNN) and long short-term memory (LSTM), a CNN-LSTM model is proposed to make accurate and point-wise prediction. To achieve the optimal performance and explore the internal mechanism of the model, structural configurations and inner states are extensively studied. CNN-LSTM can take advantage of the powerful feature extraction capacity of CNN and LSTM, and outperforms the fully-connected neural network and the plain LSTM on the experimental data of a high-speed railway. In detail, CNN-LSTM has superior performance in predicting vertical vehicle-body vibration below 10 Hz and lateral vehicle-body vibration below 1 Hz. Moreover, analysis shows that the predicted vehicle-body acceleration can act as a performance-based evaluation index of track quality.
Authors
Shuai Ma
School of Civil Engineering, Beijing Jiaotong University, Beijing, China
Other publications
>>
Liang Gao
School of Civil Engineering, Beijing Jiaotong University, Beijing, China
Other publications
>>
Xiubo Liu
Infrastructure Inspection Research Institute, China Academy of Railway Sciences Corporation Limited, Beijing, China
Other publications
>>
Jing Lin
Luleå tekniska universitet; Drift, underhåll och akustik
Other publications
>>
Record metadata
Click to view metadata
header:
identifier: oai:DiVA.org:ltu-77232
datestamp: 2021-04-19T12:42:12Z
setSpec: SwePub-ltu
metadata:
mods:
@attributes:
version: 3.7
recordInfo:
recordContentSource: ltu
recordCreationDate: 2019-12-20
identifier:
http://urn.kb.se/resolve?urn=urn:nbn:se:ltu:diva-77232
10.1109/ACCESS.2019.2960537
2-s2.0-85077963564
titleInfo:
@attributes:
lang: eng
title: Deep Learning for Track Quality Evaluation of High-Speed Railway Based on Vehicle-Body Vibration Prediction
abstract: Track quality evaluation is fundamental for track maintenance. Around the world track geometry standards are established to evaluate track quality. However these standards may not be capable of detecting some abnormal track geometry conditions that can cause considerable vehicle-body vibration. And people gradually realized that track quality evaluation should be based not only on track geometry but also on vehicle performance. Vehicle-body vibration prediction is beneficial for locating potential track geometry defects and the predicted accelerations can be used as an auxiliary index for assessing track quality. For this purpose this paper gives a method to predict vehicle-body vibration based on deep learning which represents one of the newest areas in artificial intelligence. By integrating convolutional neural network (CNN) and long short-term memory (LSTM) a CNN-LSTM model is proposed to make accurate and point-wise prediction. To achieve the optimal performance and explore the internal mechanism of the model structural configurations and inner states are extensively studied. CNN-LSTM can take advantage of the powerful feature extraction capacity of CNN and LSTM and outperforms the fully-connected neural network and the plain LSTM on the experimental data of a high-speed railway. In detail CNN-LSTM has superior performance in predicting vertical vehicle-body vibration below 10 Hz and lateral vehicle-body vibration below 1 Hz. Moreover analysis shows that the predicted vehicle-body acceleration can act as a performance-based evaluation index of track quality.
subject:
@attributes:
lang: eng
authority: uka.se
topic:
Engineering and Technology
Civil Engineering
Other Civil Engineering
@attributes:
lang: swe
authority: uka.se
topic:
Teknik och teknologier
Samhällsbyggnadsteknik
Annan samhällsbyggnadsteknik
@attributes:
lang: eng
topic: Track quality evaluation
@attributes:
lang: eng
topic: track geometry
@attributes:
lang: eng
topic: vehicle-body vibration
@attributes:
lang: eng
topic: convolutional neural network (CNN)
@attributes:
lang: eng
topic: long short-term memory (LSTM)
@attributes:
lang: eng
topic: CNN-LSTM
@attributes:
lang: swe
authority: ltu
topic: Drift och underhållsteknik
genre: Research subject
@attributes:
lang: eng
authority: ltu
topic: Operation and Maintenance
genre: Research subject
language:
languageTerm: eng
genre:
publication/journal-article
ref
note:
Published
4
Validerad;2020;Nivå 2;2020-02-17 (johcin)
name:
@attributes:
type: personal
namePart:
Ma
Shuai
role:
roleTerm: aut
affiliation: School of Civil Engineering Beijing Jiaotong University Beijing China
@attributes:
type: personal
namePart:
Gao
Liang
role:
roleTerm: aut
affiliation: School of Civil Engineering Beijing Jiaotong University Beijing China
@attributes:
type: personal
namePart:
Liu
Xiubo
role:
roleTerm: aut
affiliation: Infrastructure Inspection Research Institute China Academy of Railway Sciences Corporation Limited Beijing China
@attributes:
type: personal
authority: ltu
namePart:
Lin
Jing
role:
roleTerm: aut
affiliation:
Luleå tekniska universitet
Drift underhåll och akustik
nameIdentifier:
linjan
0000-0002-7458-6820
originInfo:
dateIssued: 2020
publisher: IEEE
relatedItem:
@attributes:
type: host
titleInfo:
title: IEEE Access
identifier: 2169-3536
part:
detail:
@attributes:
type: volume
number: 7
extent:
start: 185099
end: 185107
location:
url: https://doi.org/10.1109/ACCESS.2019.2960537
accessCondition: gratis
physicalDescription:
form: print
typeOfResource: text