集成决策解释模型和交叉影响矩阵的岩溶塌陷隐患区燃气管道安全风险分析及应用

Analysis and application of safety risks for gas pipelines in karst sinkhole-prone areas based on the D/I-MICMAC-VS integrated method

为降低岩溶塌陷隐患区燃气管道的风险水平,基于人、物、环境、管理4类事故诱因,选用DEMATEL/ISM法厘清系统因素间的层次结构和因果关系,结合MICMAC法分析风险因素的依赖度和驱动力,并基于Visual Studio平台开发了“岩溶塌陷隐患区燃气管道风险分析软件”,形成了一种D/I-MICMAC-VS集成风险分析方法并开展实例研究。结果表明:(1)岩溶塌陷隐患区燃气管道风险因素分布于6个层级,通过对表层直接因素的严格管理可以在短期内降低风险事故发生的可能性;中层间接因素在系统中起承上启下作用;只有重视深层根本因素,才能够从根本上对燃气管道事故进行控制。(2)自发集群是燃气管道事故风险管控的关键抓手,通过优先干预可对事故防治起到显著作用;独立集群通过自身的变化发展直接影响系统的风险水平;联动集群对事故的演化发展起到传递推动作用。只有厘清诱发依赖集群变化的深层根本因素才能实现有效风险管控。

To mitigate the risk of gas pipelines in karst sinkhole-prone areas,this study employs the DEMATEL/ISM method to elucidate the hierarchical structure and causal relationships among various factors in the system,considering four categories of accident causes:human,material,environment and management.Additionally,the MICMAC method is utilized to analyze the dependence and driving force of risk factors.Utilizing the Visual Studio platform,the software for risk analysis of gas pipelines in karst sinkhole-prone areas is developed.This research introduces the D/I-MICMAC-VS integrated risk analysis method and provides an example analysis.The results demonstrate that:(1)The risk factors for gas pipelines in karst sinkhole-prone areas are distributed across six levels.The possibility of risk accidents can be reduced in the short term by rigorously managing surface-level direct factors,while middle-level indirect factors play an intermediary role in the system.Effective control of gas pipeline accidents can only be achieved by addressing deep-rooted factors fundamentally.(2)The spontaneous cluster serves as a key element for risk management and control of gas pipeline accidents,and prioritized intervention significantly aids in accident prevention.The independent cluster directly influences the system’s risk level through its own changes and development.The linkage cluster plays a pivotal role in transmitting and promoting the evolution and development of accidents.Effective risk management and control can be achieved by discerning the deep root factors that inducing changes in the dependency cluster.