Data-driven approaches for modeling train control models: Comparison and case studies. (March 2020)
- Record Type:
- Journal Article
- Title:
- Data-driven approaches for modeling train control models: Comparison and case studies. (March 2020)
- Main Title:
- Data-driven approaches for modeling train control models: Comparison and case studies
- Authors:
- Yin, Jiateng
Su, Shuai
Xun, Jing
Tang, Tao
Liu, Ronghui - Abstract:
- Abstract: In railway systems, the train dynamics are usually affected by the external environment (e.g., snow and wind) and wear-out of on-board equipment, leading to the performance degradation of automatic train control algorithms. In most existing studies, the train control models were derived from the mechanical analyzation of train motors and wheel-track frictions, which may require many times of field trials and high costs to validate the model parameters. To overcome this issue, we record the explicit train operation data in Beijing Metro within three years and develop three data-driven approaches, involving a linear regression-based model (LAM), a nonlinear regression-based model (NRM), and furthermore a deep neural network based (DNN) model, where the LAM and NRM can act as benchmarks for evaluating DNN. To improve the training efficiency of DNN model, we especially customize the input and output layers of DNN, batch normalization based layers and network parameter initialization techniques according to the unique characteristics of railway train models. From the model training and testing results with field data, we observe that DNN significantly enhances the predicting accuracy for the train control model by using our customized network structure compared with LAM and NRM models. These data-driven approaches are successfully applied to Beijing Metro for designing efficient train control algorithms. Highlights: Three data-driven approaches are proposed forAbstract: In railway systems, the train dynamics are usually affected by the external environment (e.g., snow and wind) and wear-out of on-board equipment, leading to the performance degradation of automatic train control algorithms. In most existing studies, the train control models were derived from the mechanical analyzation of train motors and wheel-track frictions, which may require many times of field trials and high costs to validate the model parameters. To overcome this issue, we record the explicit train operation data in Beijing Metro within three years and develop three data-driven approaches, involving a linear regression-based model (LAM), a nonlinear regression-based model (NRM), and furthermore a deep neural network based (DNN) model, where the LAM and NRM can act as benchmarks for evaluating DNN. To improve the training efficiency of DNN model, we especially customize the input and output layers of DNN, batch normalization based layers and network parameter initialization techniques according to the unique characteristics of railway train models. From the model training and testing results with field data, we observe that DNN significantly enhances the predicting accuracy for the train control model by using our customized network structure compared with LAM and NRM models. These data-driven approaches are successfully applied to Beijing Metro for designing efficient train control algorithms. Highlights: Three data-driven approaches are proposed for predicting the train control models. An improved DNN model is developed. The models are trained and tested with real-world data. The models are implemented in constructing a virtual platform for supporting the real-world applications in Beijing Metro. … (more)
- Is Part Of:
- ISA transactions. Volume 98(2020)
- Journal:
- ISA transactions
- Issue:
- Volume 98(2020)
- Issue Display:
- Volume 98, Issue 2020 (2020)
- Year:
- 2020
- Volume:
- 98
- Issue:
- 2020
- Issue Sort Value:
- 2020-0098-2020-0000
- Page Start:
- 349
- Page End:
- 363
- Publication Date:
- 2020-03
- Subjects:
- Train control models -- Data-driven approaches -- Artificial neural networks -- Field test
Engineering instruments -- Periodicals
Engineering instruments
Periodicals
Electronic journals
629.805 - Journal URLs:
- http://www.sciencedirect.com/science/journal/00190578 ↗
http://www.elsevier.com/journals ↗ - DOI:
- 10.1016/j.isatra.2019.08.024 ↗
- Languages:
- English
- ISSNs:
- 0019-0578
- Deposit Type:
- Legaldeposit
- View Content:
- Available online (eLD content is only available in our Reading Rooms) ↗
- Physical Locations:
- British Library DSC - 4582.700000
British Library DSC - BLDSS-3PM
British Library HMNTS - ELD Digital store - Ingest File:
- 13353.xml