论文部分内容阅读
建立了一套基于预测的公交信号优先干线联动控制方法。考虑到公交车辆在停靠站的停留时间受到多重因素的综合影响,首先构建了ARIMA-SVR的组合模型用于预测公交车辆的站点延误,并以此为重要依据,预测了公交到达交叉口的时间。通过比较车辆预计到达时间与理想时间的差值,计算了延伸和压缩信号周期的惩罚因子,根据惩罚因子的大小调整了信号周期和绿灯时间。在进行交叉口间协同控制时,又将交叉口平面设计的物理条件和交叉口群的协同控制条件纳入对信号的调整进行约束。为了验证该方法的实际应用效果,为某城市快速公交工程实例设计了VISSIM信号优先模块。研究结果表明:组合模型预测公交车辆站点延误的相关系数为0.890 4,预测精度较好;在改进的信号优先算法的情况下,公交交叉口延误比现状降低近50%,公交车车头时距一致性平均下降38.8%,且该算法为信号调整提供了较为充足的缓冲空间,在绿灯时间调整时兼顾考虑了对社会车辆的影响,因此,在预测式信号优先中,社会车辆的行驶延误和交叉口排队长度也较其他优先方法有所降低。
A set of prediction-based bus priority linkage control method is established. Taking into account the bus stop time in the bus station by a combination of multiple factors, the first ARIMA-SVR combination model is constructed to predict the delay of the bus station site, and as an important basis for predicting the time to reach the intersection of the bus . By comparing the difference between expected arrival time and ideal time, the penalty factor of extended and compressed signal period is calculated, and the signal period and green light time are adjusted according to the size of penalty factor. In the process of coordinated control between intersections, the physical conditions of the intersection graphic design and the collaborative control conditions of the intersection group are also included in the adjustment of the signal. In order to verify the practical application of this method, a VISSIM signal priority module is designed for a city bus rapid transit project. The results show that the combined model predicts the delay of bus stops is 0.890 4 and the prediction accuracy is better. In the case of improved signal priority algorithm, the delay of bus intersections is reduced by nearly 50% The average decrease of 38.8%, and the algorithm for the signal adjustment provides ample buffer space, taking into account the impact on social vehicles in the green light time adjustment, therefore, in predictive signal priority, the social vehicle delay and cross Port queue length is also lower than other priority methods.