Short-term Traffic Prediction Model Based on Grey Neural Network

摘要

This paper expounds three kinds of grey neural network combined model for short-term prediction of urban traffic speed, and confirms their validity and feasibility by conducting experiment in Beijing road of Jingzhou. Three kinds of networks are parallel grey neural network, series grey neural network, and inlaid grey neural network. The experiment proves that the three kinds of modes are feasible and effective in comparison with single model GM(1, 1) and neural network. And actual traffic speed varies smoothly or will not influence significantly the accuracy for prediction.