为了实现利用视频车辆检测器数据计算和预测路段行程时间,将排队长度数据应用到路段行程时间的计算中,采用改进粒子群的BP神经网络算法和时间序列分析对路段进行实证研究.将排队长度加入计算得到的决定系数为93.36%,比只有流量数据的BP...为了实现利用视频车辆检测器数据计算和预测路段行程时间,将排队长度数据应用到路段行程时间的计算中,采用改进粒子群的BP神经网络算法和时间序列分析对路段进行实证研究.将排队长度加入计算得到的决定系数为93.36%,比只有流量数据的BP神经网络算法改善了41.03%,比BPR(bureau of public roads)路阻函数算法改善了23.37%.利用实时的路段行程时间对后续行程时间预测通过时间序列分析得到相对误差为0.06,预测下个时段和下个周期的路段行程时间平均相对误差分别为0.14、0.15.结果表明排队长度对于路段行程时间的计算具有较高的准确性,可以用于城市道路交通时间的预测,并能有效为智能交通算法的其他指数计算提供思路,为改善交通状况提供决策支持.展开更多
Fluid flow throttling is common in industrial and building services engineering.Similar tunnel throttling of vehicular flow is caused by the abrupt number reduction of roadway lane,as the tunnel has a lower lane numbe...Fluid flow throttling is common in industrial and building services engineering.Similar tunnel throttling of vehicular flow is caused by the abrupt number reduction of roadway lane,as the tunnel has a lower lane number than in the roadway normal segment.To predict the effects of tunnel throttling of annular freeway vehicular flow,a three-lane continuum model is developed.LaneⅢof the tunnel is completely blocked due to the need of tunnel rehabilitation,etc.There exists mandatory net lane-changing rate from laneⅢto laneⅡjust upstream of the tunnel entrance,which is described by a model of random number generated through a golden section analysis.The net-changing rate between adjacent lanes is modeled using a lane-changing time expressed explicitly in algebraic form.This paper assumes that the annular freeway has a total length of 100 km,a two-lane tunnel of length 2 km with a speed limit of 80 km/h.The free flow speeds on lanesⅠ,ⅡandⅢare assumed to be 110,100 and 90 km/h respectively.Based on the three-lane continuum model,numerical simulations of vehicular flows on the annular freeway with such a tunnel are conducted with a reliable numerical method of 3rd-order accuracy.Numerical results reveal that the vehicular flow has a smaller threshold of traffic jam formation in comparison with the case without tunnel throttling.Vehicle fuel consumption can be estimated by interpolation with time averaged grid traffic speed and an assumed curve of vehicle performance.The vehicle fuel consumption is lane number dependent,distributes with initial density concavely,ranging from 5.56 to 8.00 L.Tunnel throttling leads to an earlier traffic jam formation in comparison with the case without tunnel throttling.展开更多
文摘为了实现利用视频车辆检测器数据计算和预测路段行程时间,将排队长度数据应用到路段行程时间的计算中,采用改进粒子群的BP神经网络算法和时间序列分析对路段进行实证研究.将排队长度加入计算得到的决定系数为93.36%,比只有流量数据的BP神经网络算法改善了41.03%,比BPR(bureau of public roads)路阻函数算法改善了23.37%.利用实时的路段行程时间对后续行程时间预测通过时间序列分析得到相对误差为0.06,预测下个时段和下个周期的路段行程时间平均相对误差分别为0.14、0.15.结果表明排队长度对于路段行程时间的计算具有较高的准确性,可以用于城市道路交通时间的预测,并能有效为智能交通算法的其他指数计算提供思路,为改善交通状况提供决策支持.
基金supported by the project of National Natural Science Foundation of China“exploring the road condition effect of travel time using emergency mitigation traffic flow models”(grant 11972341)fundamental research project of Lomonosov Moscow State University“mathematical models for multi-phase media and wave processes in natural,technical and social systems”。
文摘Fluid flow throttling is common in industrial and building services engineering.Similar tunnel throttling of vehicular flow is caused by the abrupt number reduction of roadway lane,as the tunnel has a lower lane number than in the roadway normal segment.To predict the effects of tunnel throttling of annular freeway vehicular flow,a three-lane continuum model is developed.LaneⅢof the tunnel is completely blocked due to the need of tunnel rehabilitation,etc.There exists mandatory net lane-changing rate from laneⅢto laneⅡjust upstream of the tunnel entrance,which is described by a model of random number generated through a golden section analysis.The net-changing rate between adjacent lanes is modeled using a lane-changing time expressed explicitly in algebraic form.This paper assumes that the annular freeway has a total length of 100 km,a two-lane tunnel of length 2 km with a speed limit of 80 km/h.The free flow speeds on lanesⅠ,ⅡandⅢare assumed to be 110,100 and 90 km/h respectively.Based on the three-lane continuum model,numerical simulations of vehicular flows on the annular freeway with such a tunnel are conducted with a reliable numerical method of 3rd-order accuracy.Numerical results reveal that the vehicular flow has a smaller threshold of traffic jam formation in comparison with the case without tunnel throttling.Vehicle fuel consumption can be estimated by interpolation with time averaged grid traffic speed and an assumed curve of vehicle performance.The vehicle fuel consumption is lane number dependent,distributes with initial density concavely,ranging from 5.56 to 8.00 L.Tunnel throttling leads to an earlier traffic jam formation in comparison with the case without tunnel throttling.
