自适应神经网络交通流预测的混沌麻雀优化方法  

Optimizing Neural Network Based on Chaotic Sparrow for Traffic Flow Forecasting

作　　者：王敬月 楼俊钢[1,2] WANG Jingyue;LOU Jungang(School of Information Engineering,Huzhou University,Huzhou 313000,China;Zhejiang Province Key Laboratory of Smart Management&Application of Modern Agricultural Resources,Huzhou 313000,China)

机构地区：[1]湖州师范学院湖州长三角智慧交通研究院,浙江湖州313000 [2]浙江省现代农业资源智慧管理与应用研究重点实验室,浙江湖州313000

出　　处：《湖州师范学院学报》2022年第10期52-59,共8页Journal of Huzhou University

基　　金：国家自然科学基金项目(61772199);浙江省自然科学基金项目(LR20F020002)。

摘　　要：群体优化神经网络模型可以以任意精度逼近各种复杂函数,在各类非线性预测问题包括交通流预测中得到了很好的应用.鉴于其局部寻优能力强、收敛速度快等特点,采用麻雀算法优化自适应神经网络交通流预测模型的参数,并在麻雀算法中引入混沌映射,以增加其搜索的遍历均匀性和随机性,从而优化其全局寻优性能.多个实际交通流数据集的实验结果表明,相比常用的群体优化神经网络模型,优化后的自适应神经网络交通流预测模型的平均预测准确度提升0.52%~4.01%.The swarm optimization neural network model can approximate various complex functions with arbitrary accuracy, and has been well applied in various nonlinear prediction problems including traffic flow prediction. In view of its strong local optimization ability and fast convergence speed, the sparrow algorithm is used to optimize the parameters of the adaptive neural network traffic flow prediction model;chaotic mapping is introduced in the sparrow algorithm to increase the uniformity and randomness of the search traversal, thereby optimizing the overall situation Optimal performance. Experimental results on multiple actual traffic flow data sets show that the method proposed in this paper improves the average prediction accuracy by 0.52%~4.01% compared with the commonly used swarm intelligence optimization neural network method.

关 键 词：智能交通 麻雀算法 TENT映射 参数优化 交通流预测 

分 类 号：U491[交通运输工程—交通运输规划与管理]