Resilience Identification And Assessment Of Urban Subway Station System To Fire Disaster

As the complexity and diversity of disturbances such as natural disasters,how to enhance the safety of lifeline infrastructure system operation has become the focus of urban disaster prevention and mitigation.Under this background,the Safety-Ⅱ system safety research paradigm with resilience as its core,provides new ideas for urban disaster response from the perspective of dynamics and development.Among all infrastructures,subway is an effective tool to mitigate urban traffic congestion and promote urban development,and subway stations,as train stops and passenger pick-up and drop-off place,are the core part of subway operation.Quantities of subway accidents have shown that subway station system fire accidents are the most important safety accidents in the safety management of subway station systems.Therefore,this paper takes subway station system-fire as a binary research object and studies the identification,measurement and improvement of the fire resistance of the subway station.The main contents and results of this paper are as follows.Firstly,the definition of fire resistance is proposed based on System Resilience Theory.Fire resilience formation mechanism analysis of subway station system was analyzed based on Resilience TOSE-4Rs Framework.Fire resilience emergence mechanism analysis of subway station system was analyzed based on Catastrophe Dynamics.Secondly,method based on resilience ability was adopted to evaluate fire resilience from the perspectives of indicators and scenarios.Based on standards,cases and literatures identification method,36 factors were identified that would have an influence on fire resilience.Based on Question Survey and the improved TOPSIS,key influencing factors were screened out.Thirdly,combined with expert experience,causal relationship between factors were analyzed and fire resilience Bayesian Network of subway station system was built.Based on Fuzzy Comprehensive Evaluation method,the prior probability of the root nodes was obtained.Based on Leaky Noisy-OR model,the conditional probability table for non-root nodes was obtained.Netica software was used to make forward and backward propagation analysis in order to evaluate fire resilience and obtain the most important cause chain that lead to fire resilience failure.Finally,based on Bayesian Network and Sensitivity Analysis,influencing factors that have a great impact on fire resilience were identified and a static promotion strategy for fire resilience was proposed.Based on Dynamic Bayesian Network and Importance Analysis,influencing factors that contribute significantly to reducing the probability of failure of fire resilience(probability importance),have a higher probability of triggering leaf node failure(critical importance),and have a great impact on the fire resilience when it completely failed(structural importance)were identified and a dynamic promotion strategy for fire resilience was proposed.This paper introduces the concept of resilience in the safety management of fire accidents in subway station systems.Based on system resilience theory,this paper comprehensively explores the full process identification,scene-oriented measurement and dynamic improvement research method for the fire resilience of the subway station system in the fire life cycle and proposes a full-process resilient management idea as prevention-emergency-recovery-feedback for subway station operators.