New Requirements Elicited from Accidents of Train Control System Simulated by Modelica.

Wen Zhong,Yan Wang,Xiaohong Chen
DOI: https://doi.org/10.1007/978-981-10-7796-8_15
2018-01-01
Abstract:Although train control system (TCS) is regarded relatively safe, accidents still happened from time to time. In this paper, we propose a simulation based approach to elicit new requirements from accidents and then modify the TCS to provide a more reliable and safer system. A Modelica system model is constructed to describe the structure and interactions of TCS according to the continuous behavior and discrete fault event of TCS devices. A Modelica accident model is also defined based on the system model in order to predict accidents. These Modelica models are simulated in Open Modelica until all scenarios (paths) are covered. By analyzing the simulation results which indicate the causes of accidents, we elicit new requirements, and modify the original system model. Simulation is used again to show that these modifications could effectively avoid such accidents. A case study is provided to validate our approach.
What problem does this paper attempt to address?