论文部分内容阅读
隧道是城市交通的咽喉所在,为了有效引导隧道交通流,实现隧道交通状况的实时发布,从而引导驾驶员理性进出隧道,准确可靠的数据采集及实时交通流预测是关键所在。以VISSIM仿真为基础,建立研究隧道路网模型,以行程时间误差为依据,探讨了检测器在隧道中的空间合理布局,保证数据采集的可靠性。在准确的数据采集基础上,选取了自适应权重指数平滑法(AWES)以及基于径向基函数的神经网络模型(RBFN)进行实证研究,并采用数据融合技术以保障预测精度及误差的稳定性。并进一步以VISSIM仿真平台简略探讨了交通状态判别指标,以期更好服务于城市交通诱导系统。
Tunnel is the throat of urban traffic. In order to effectively guide the traffic flow in the tunnel and realize the real-time release of traffic conditions in the tunnel, it is essential to guide the driver to enter and leave the tunnel rationally. Accurate and reliable data acquisition and real-time traffic flow prediction are the keys. Based on the VISSIM simulation, a tunnel network model is established. Based on the travel time error, the spatial distribution of the detector in the tunnel is discussed to ensure the reliability of data acquisition. On the basis of accurate data collection, we choose adaptive weighted index smoothing (AWES) and neural network model based on radial basis function (RBFN) for empirical research, and use data fusion technology to ensure the prediction accuracy and the stability of the error . Furthermore, the VISSIM simulation platform is also used to briefly discuss the traffic status discrimination index in order to better serve the urban traffic guidance system.