基于改进沙漏模型的突发事件下交通流预测

通过分析突发事故导致车道被占用时，道路通行能力的演变过程及交通流的 变化特征，将占道发生后车流与沙漏模型中颗粒物质运动类比，结合突发事件下交通流 中不同类型车辆的换道规律，提出了含概率崩塌各异性的改进沙漏模型.并结合元胞自动 机仿真理论，运用MATLAB进行仿真计算不同时刻的车辆排队长度，与实际数据对比， 该模型的平均相对误差为6.509 7%，验证了模型的可靠性.最后利用该模型预测不同车道 被占用和不同车流量的情况下车队长度达到特定长度所需的时间，进而探讨其分别对道 路通行能力的不同影响程度，为交通部门监管道路提供理论依据.

The Forecasting of Traffic Flow Based on Optimized Hourglass Model in Emergencies

From the analysis of the evolving of traffic flow when traffic congestion emerges caused by accident, analogies of traffic flow with the movement of particulate matter in hourglass model and various vehicles’lane changing pattern in emergency situation, an optimized hourglass model is propose. Based on cellular automata method, we simulate the queuing lengths of vehicles in different time on MATLAB and compare them with the practical data .We find that the average relative error of the model is 6.509 7%, verifying the reliability of the model. Finally, according to the model, we predict the time required for the specific queuing length on various lanes and traffic volumes. Thusly, we can study the impacts of time on traffic capacity and provide a theoretical basis for transportation administration.