基于贝叶斯网络的高速列车碰撞风险评价  

Study on High-speed Train Collision Risk Assessment Based on Bayesian Network

作　　者：罗军[1] 李颖[2] 马宏锋[2] 

机构地区：[1]西安铁路局西安供电段,陕西西安710000 [2]兰州工业学院电子信息工程系,甘肃兰州730050

出　　处：《兰州工业高等专科学校学报》2013年第2期1-5,共5页Journal of Lanzhou Higher Polytechnical College

基　　金：铁道部科技计划项目(2010G014-G);甘肃省自然科学基金(096RJZA084)

摘　　要：碰撞事故已成为影响高速列车运营安全的重要因素,采用科学方法对安全风险进行有效的预测与评估具有重要意义.从组织、人因错误的角度出发,利用贝叶斯网络构建了碰撞事故的致因关系.参考专家知识,基于风险发生概率语言变量量化基本事件的条件概率,构建条件概率表,利用GeNIe软件进行了概率推理和风险因素灵敏度分析,定量评估了列车碰撞的风险,找出了影响最突出的风险因素.通过案例研究表明,贝叶斯网络对于高速列车运营安全风险预测是可行的,计算得到的风险概率不仅指出了重要的影响因素,也反映了高速列车系统目前的安全状况.The collision has become an important factor to the safety of high-speed train operation, and it is im- portant to apply scientific and effective method to predict and assess security risk. From the perspectiue of organi- zation and human error, collision cause relationship is constructed by Bayesian network. With reference to expert knowledge, the basic event probability is quantified based on the probability of occurrence of the risk of linguistic variables; conditional probability table is built; probabilistic reasoning and sensitivity analysis of the risk factors is realized by GeNie; quantitative assessment of the train collision risk is conducted; the impact of the most prominent risk factors is found out. Case studies show that the Bayesian network is fit for security risk prediction of high-speed train. Risk probability not only points out the important factor, but also reflects the current security situation in high-speed train system.

关 键 词：高速列车 贝叶斯网络 风险评价 灵敏度分析 

分 类 号：TP18[自动化与计算机技术—控制理论与控制工程]