Evaluating Fault Tree by means of Colored Petri nets to analyze the railway system dependability

摘要

Railway system is a safety critical and time-related system, the system's states and time parameters can be used to carry out the dependability and hazard analysis. Fault Tree is widely recognized as a standard evaluating method. However, restricted by the commercial products, the Fault Tree is limited to assess dynamic systems with event-repair operations and time-related attributions. Additionally, it is difficult to incorporate non-linear relationships such as feedback. The quality assurance for fault trees and events trees is mainly carried out by peer review. Combinatory limitations are encountered when modeling complex events with classical methods. Thus, this paper proposes a new method to represent and extend the Fault Tree in Colored Petri nets. Due to large calculation capabilities of CPNs, these limitations can be able to overcome. Additionally, it can be reused for customizations. The accuracy of the approach is verified by using model-based simulation and state space analysis. The performance and benefits of the new approach are demonstrated by investigating train to train collision failure models. To increase the safety demanding needs of railway transportation, we propose a new train movement authority plus system (MA +) in this paper. With the assistance of the wireless communication technology, MA + can detect the condition of approaching switches and encountering trains within a certain range. The results indicate that the new MA + can reduce the risk of train head to tail collisions. What is more, the new evaluation method can offer much more essential information, which involves maintenance components, model correctness verification, time factors, and mathematical calculation together, than the traditional Fault Tree Analysis.