基于改进的遗传算法优化的BP神经网络交通流量预测模型

Traffic Flow Prediction Model of BP Neural Network Based on Adaptive Genetic Algorithm Optimization

摘要: 短时交通流量是一种具有较强的随机性数据，不仅受到上下游历史数据的影响，而且还与天气等因素有关。为了提高短时交通流量预测的精度，采用遗传算法与BP神经网络组合的方法进行研究。同时针对BP神经网络存在网络初始权值随机的缺陷以及传统遗传算法易陷入局部最优和过早收敛等问题，给出了一种基于自适应遗传算法优化的BP神经网络短时交通流量预测模型，该模型通过自适应遗传算法对BP神经网络的初始权值进行优化，从而对短时交通流量进行预测分析。并结合实验数据验证了该模型的可靠性，该模型可以优化出行体验以及为交通管理部门制定缓解交通拥堵措施。

Abstract: Short-term traffic flow is a kind of strong random data, which is not only affected by historical data of upstream and downstream, but also related to weather factors. In order to improve the accuracy of short-term traffic flow prediction, a combination of genetic algorithm and BP neural network is used for research. At the same time, aiming at the shortcomings of the random initial weight of the BP neural network and the traditional genetic algorithm is easy to fall into local optimum and premature convergence, a short-term traffic flow prediction of the BP neural network based on the adaptive genetic algorithm optimization is given A model that optimizes the initial weight of a BP neural network through an adaptive genetic algorithm to predict and analyze short-term traffic flow. The reliability of the model is verified with experimental data. The model can optimize the travel experience and specify traffic congestion measures for the traffic management department.