Road capacity calibration of medium and small cities based on radial basis function neural network

Abstract

Based on the speed-flow model and the principle of radial basis function (RBF) neural network, this article designs the capacity calibration method of small and medium-sized city road sections. Compared with the traditional method, which gets the road capacity by modifying the design road capacity, the method avoids the problem of low accuracy due to the lack of consideration of random factors in the deterministic mathematical model, which is suitable for the complicated traffic environment of the road section in small and medium-sized cities. Then, this article takes Xiangshui old city road network as an example, and investigates the related attributes such as the number of lanes, the width of the red line, whether designing parking space. At the same time, the flow and speed of some road sections are investigated as training sections, then the designed method is applied to calibrate the road capacity of Xiangshui old city. The error between the predicted value and the real value of the training road sections is within the allowable range of the project, the correctness of the design method is verified, it can lay a data foundation for urban traffic planning and management.