Short-Term Traffic Flow Forecasting Based on GM-Fusion Neural Network Hybrid Model—A Case Study of Linyi City, Shandong Province

随着社会发展，人们经济水平不断提高，交通拥堵已成为制约我国城市可持续发展的重要因素之一，而交通流量预测是智能交通系统中的重要环节。为解决该问题，提出优化模型，提高交通流量预测的精确度.本项目收集了山东省临沂市某主要商业圈周围道路的大量车流量数据，经数据预处理，选取交通流量预测方面表现出色的灰色GM (1, N)和神经网络融合的模型，该模型将交通流量的多种影响因素综合考虑，使用GM (1, N)模型预测天气、早晚高峰等影响因素下的交通流量，融合神经网络系统，计算权重，并且通过BP神经网络根据误差不断修正新的权重，使得误差满足精度要求，进而得出优化的交通流量预测值，提高了交通流量预测模型的精确度。With the development of society and the continuous improvement of people’s economic level, traffic congestion has become one of the important factors restricting the sustainable development of cities in China. Traffic flow prediction is an important part of the intelligent transportation system. To solve this problem, an optimization model is proposed to improve the accuracy of traffic flow prediction. This project collected a large amount of traffic flow data from the roads around a major commercial circle in Linyi City, Shandong Province. After data preprocessing, a model that combines the outstanding grey GM (1, N) and neural network in traffic flow prediction was selected. This model comprehensively considers various influencing factors of traffic flow, uses the GM (1, N) model to predict traffic flow under influencing factors such as weather and rush hours, integrates the neural network system, calculates the weights, and continuously corrects the new weights through the BP neural network based on the error, so that the error meets the accuracy requirements, and then obtains the optimized traffic flow prediction value, improving the accuracy of the traffic flow prediction model.