基于动态贝叶斯网络的道路交通系统抗震韧性评价

Seismic Resilience Evaluation of the Traffic System Based on the Dynamic Bayesian Network

以道路交通系统在地震作用下的连通性随时间变化的曲线作为其抗震韧性量化评价的基础,构建了基于动态贝叶斯网络的抗震韧性评价模型。该模型采用动态贝叶斯网络表征道路交通系统抗震功能的连通逻辑以及路段连通性随时间的变化趋势,并基于网络中各变量的概率与条件概率表输出不同时刻系统的连通性,绘制连通性曲线。最后以青岛市的局部城市路网为应用实例,实现了对道路交通系统全过程抗震韧性的评价。结果表明:所提出的韧性指标可全面描述震后系统连通性退化与恢复全过程;构建的动态贝叶斯网络能直观表征系统抗震功能的连通逻辑与变化;全过程韧性评价结果能够为系统震前预防与震后恢复策略的制定提供参考。

To construct the resilience evaluation model,the time-varying connectivity curve of the traffic system hit by an earthquake event is taken as the basis for the quantitative evaluation of its seismic resilience.The Dynamic Bayesian Network(DBN)is used to characterize the connectivity logic of the seismic function of road traffic system and the change trend of link connectivity with time.Based on the probability of each variable and conditional probability tables in the network,the connectivity of the system at different times is output and the connectivity curve is drawn.Taking the local urban road network in Qingdao City as an example,the whole process evaluation of the seismic resilience of the road traffic system is realized.The results show that the resilience index proposed in this paper can comprehensively describe the whole process of system’s connectivity degradation and recovery after an earthquake.The constructed DBN can intuitively characterize the connection logic and change of seismic function of the system.The whole process resilience evaluation results can provide reference for the formulation of earthquake preparedness and post-earthquake recovery strategies.