文献类型：期刊文献

中文题名：针对短时交通流预测的ISSA-SVR模型

英文题名：Improved sparrow search algorithm optimizing support vector regression model for short-term traffic flow prediction

作者：叶得学[1];韩如冰[1];颜鲁合[2]

第一作者：叶得学

机构：[1]兰州工商学院信息工程学院,甘肃兰州730101;[2]甘肃中医药大学经贸与管理学院,甘肃兰州730101

第一机构：兰州工商学院信息工程学院,甘肃兰州730101

年份：2024

卷号：45

期号：2

起止页码：608

中文期刊名：计算机工程与设计

外文期刊名：Computer Engineering and Design

收录：CSTPCD;;北大核心:【北大核心2023】；

基金：甘肃省科技计划基金项目(20CX9ZA021);甘肃省高等学校创新基金项目(2020A-175);兰州市科技计划基金项目(2020-ZD-139);甘肃省科技计划基金项目(20CX9ZA068)。

语种：中文

中文关键词：交通流预测;智能交通;麻雀搜索算法;支持向量回归;反向学习;蝴蝶优化算法;柯西变异

外文关键词：traffic flow prediction;intelligent transportation;sparrow search algorithm;support vector regression;opposite learning;butterfly optimization algorithm;Cauchy mutation

摘要：为提高短时交通流的预测精度,提出一种改进麻雀搜索算法优化支持向量回归的预测模型。为解决麻雀搜索算法收敛慢、易陷入局部最优的不足,结合反向学习和中心游移进行种群初始化;引入分段惯性权重和蝴蝶优化算法改进发现者更新,扩展迭代早期的全局搜索范围和寻优能力;利用柯西变异追随者更新机制提高迭代后期的局部开发能力和收敛速度;设计自适应警戒者更新均衡搜索和开发过程。应用改进麻雀搜索算法优化支持向量回归模型,构建短时交通流预测模型HMSSSA-SVR。实验结果表明,改进模型的泛化能力更好,预测误差更低,能够对短时交通流实现精确预测。
To improve the accuracy of short-term traffic flow prediction,a prediction model based on improved sparrow search algorithm and optimized support vector regression was proposed.To solve the problems of slow convergence and easiness to fall into local optimum of SSA,the population initialization was combined with reverse learning and center walk mechanism to improve the population diversity.The segmented inertia weight and the butterfly optimization algorithm were introduced to improve the discoverer’s position and expand the global search range and optimization ability in the early iteration of the algorithm.A follower location update mechanism based on Cauchy mutation was proposed to improve the local development ability and convergence speed of the algorithm in the later stage of iteration.An adaptive vigilant location update mechanism was designed to balance the search and development and improve the optimization accuracy of the algorithm.The improved sparrow search algorithm was applied to optimize SVR model,and a short-term traffic flow prediction model was constructed.Experimental results show that the improved model has better generalization ability and lower prediction error,and it can accurately predict the short-term traffic flow.