摘要
针对小波网络结构不稳定和容易陷入局部最小造成预测结果误差过大的问题,以辽宁省某高速公路为研究对象,利用遗传算法具有自适应随机优化搜索能力、收敛速度快的特点,提出一种基于GA-WNN神经网络的高速公路日交通流量预测方法。模型仿真结果表明,遗传算法优化小波神经网络预测的误差精度为8.35%,与传统BP神经网络和小波神经网络相比,预测精度显著提高,具有更好的预测能力。
In view of the large prediction error caused by the unstable structure of wavelet network and the tendency of falling into local minimum,an expressway daily traffic flow prediction method based on GA-WNN(genetic algorithm-wavelet neural network)is proposed by utilizing the genetic algorithm characteristics of adaptive random optimization search capability and high convergence rate.A certain expressway in Liaoning province is taken as the research object in this paper.The model simulation results show that the error accuracy predicted by genetic algorithm optimized wavelet neural network is 8.35%,which is significantly improved compared with that obtained by the traditional BP neural network and wavelet neural network.Therefore,the proposed method has a better ability of prediction.
作者
王小凡
朱永强
潘福全
WANG Xiaofan;ZHU Yongqiang;PAN Fuquan(School of Mechanical and Automotive Engineering,Qingdao University of Technology,Qingdao 266520,China)
出处
《现代电子技术》
北大核心
2020年第15期131-134,共4页
Modern Electronics Technique
基金
国家自然科学基金项目(51505244)
教育部人文社会科学研究规划基金(18YJAZH067)
山东省重点研发计划项目(2018GGX105009)。
