Computational & Technology Resources
an online resource for computational,
engineering & technology publications |
|
Civil-Comp Conferences
ISSN 2753-3239 CCC: 7
PROCEEDINGS OF THE SIXTH INTERNATIONAL CONFERENCE ON RAILWAY TECHNOLOGY: RESEARCH, DEVELOPMENT AND MAINTENANCE Edited by: J. Pombo
Paper 4.14
Continuous Fault Injection for Pantographs of High-Speed Trains Based on AMESim and Python T. Xia1, J. Ding2,1, J. Zuo1,3 and Y. Pan1
1Institute of Rail Transit, Tongji University, Shanghai, China
T. Xia, J. Ding, J. Zuo, Y. Pan, "Continuous Fault Injection for Pantographs of High-Speed Trains Based on AMESim and Python", in J. Pombo, (Editor), "Proceedings of the Sixth International Conference on
Railway Technology: Research, Development and Maintenance",
Civil-Comp Press, Edinburgh, UK,
Online volume: CCC 7, Paper 4.14, 2024, doi:10.4203/ccc.7.4.14
Keywords: digital twin, fault injection, fault diagnosis, high-speed train, pantograph, near real-time simulation.
Abstract
As the high-speed rail technology rapidly evolves, the demands for equipment safety and reliability have become more stringent.
This study introduces a novel continuous fault injection methodology that integrates AMESim simulations with Python scripting,
exemplified through an analysis of the pantograph model.
This approach provides an innovative technical means for continuous fault injection within high-speed rail systems by automating and perpetuating fault simulations.
The paper meticulously details the mechanisms of fault injection implementation and substantiates the method's efficacy and precision through case studies involving a basic model and the pneumatic system of a high-speed rail pantograph.
The experimental outcomes indicate that the continuous fault injection technique sustains the simulation's continuity and guarantees the fidelity of the simulation outcomes.
Additionally, this research delves into the integration of fault injection within digital twin platforms and highlights pertinent considerations for practical deployment.
download the full-text of this paper (PDF, 9 pages, 1136 Kb)
go to the previous paper |
|