基于贝叶斯网络推理的道路网络级联失效仿真.pdfVIP

第15卷第2 期 交通运输系统工程与信息 Vol. 15 No.2 2015 年4 月 Journal of Transportation Systems Engineering and Information Technology April 2015 文章编号：1009-6744（2015）02-0210-06 中图分类号：U491 文献标志码：A 基于贝叶斯网络推理的道路网络级联失效仿真 刘新全*a,b,c （广西师范学院 a.北部湾环境演变与资源利用教育部重点实验室；b.广西地表过程与智能模拟重点实验室； c.物流管理与工程学院，南宁530001） 摘 要： 针对道路网络级联失效与其它复杂网络相比具有的特殊性，基于贝叶斯网络 推理，融合驾驶者出行先验及出行信息，研究了道路网络级联失效的演变机理及演变过 程.编制MATLAB 程序生成基于道路网络的贝叶斯网络结构，根据贝叶斯网络参数学习 MATLAB 程序，学习了驾驶者感知路段属性变化；给出了仿真算法，仿真了贝叶斯网络推 理中不同出行先验权重，以及出行信息量对道路网络属性及级联失效的影响.结果表明， 贝叶斯网络推理可以较好地反映驾驶者出行路段选择对道路网络级联失效的定量影响， 为研究道路网络级联失效提供了新的研究思路和方法. 关键词： 交通工程；级联失效；贝叶斯网络推理；道路网络；仿真 Cascading Failures Simulation of Road Networks Based on Bayesian Network Inference LIU Xin-quana, b, c (a. Key Laboratory of Environment Change and Resources Use in Beibu Gulf ；b. Guangxi Key Laboratory of Earth Surface Processes and Intelligent Simulation ；c. College of Economics and Management, Guangxi Teachers Education University, Nanning 530001, China) Abstract:Abstract: The characteristics of road networks cascading failures are studied to other complex networks. The evolution of mechanism and process of road networks cascading failures are studied by integrating travel prior experience and travel information based on Bayesian network inference. The Bayesian network structure of road network is generated by MATLAB program. The drivers perception of road properties change is learned based on the MATLAB program of Bayesian network parameter learning. We design the simulation algorithms and simulate the ef