高速公路交通流模型是一个高阶非线性时变系统 ,这使得该模型的辨识问题成为一个非常困难的问题。简要介绍了模型及其中各参数的含义 ,在对模型参数加以分析、讨论的基础上将其分类并分别采用径向基函数 (RBF)神经网络和最小二乘法对模型的参数进行分类辨识 ,成功地解决了该模型辨识的工程化问题。通过与传统的复合形法的辨识结果进行比较 ,该方法的辨识精度和速度均明显提高 ,仿真效果令人满意 The highway traffic flow model is a high-order nonlinear time-varying system, which makes the model identification problem a very difficult problem. The model and the meaning of each parameter are briefly introduced. Based on the analysis and discussion of the model parameters, the parameters are classified and identified by RBF neural network and least square method, respectively. Solved the engineering problem of the model identification. Compared with the traditional composite method, the identification accuracy and speed of the method are obviously improved, and the simulation results are satisfactory