基于bp神经网络的快速路小时交通量预测本科毕业设计.pdf

五邑大学本科毕业设计 基于 BP 神经网络的快速路小时交通量预测 摘 要 快速路高效运行是解决城市交通拥堵问题的关键所在， 城市 ITS 的应用提高了城市路 网使用效率以及发挥环路运营管理的使用功效， 准确的快速路交通量预测数据为城市 ITS 提供数据基础。本设计应用 BP 神经网络对快速路小时交通量进行预测，通过确定 BP 神 经网络的结构和采用 MATLAB 神经网络工具箱函数建立神经网络预测模型，把已知的历 史交通量数据作为输入、 输出训练的样本集， 依据已知数据和实际输出间的误差， 用学习 系统来调节系统参数提高预测的精度， 然后使用 MATLAB 对仿真程序进行仿真得出结果， 仿真结果表明 BP 神经网络对交通量进行预测效果很好。 矚慫润厲钐瘗睞枥庑赖。 关键词： 快速路；交通量预测； BP 神经网络；仿真 I 五邑大学本科毕业设计 Abstract The efficiency of expressway is the key to solve the problem of the urban traffic congestion. Moreover, The application of ITS develops the efficiency of the urban expressway network and gives scope to the potential of the loop ’s operation management, the ITS data is based on the accurate forecast figures of expressway traffic flow. In this design, BP neural network is used to predict the traffic volume of expressway, by determining the structure of the BP neural network and adopting MATLAB neural network toolbox function to create a neural network prediction model, use known historical traffic data as the sample set of input and output training. And use learning system to adjust the system parameters to improve the prediction accuracy according to the error between the known data and the actual output. Then use MATLAB simulation program for the simulation results, the simulation results show that the BP neural network to 聞創沟燴鐺險爱氇谴净。 forecast the traffic volume effect is very good. 残骛楼諍锩瀨濟溆塹籟。 Ke