基于CS优化RVM的短时交通流预测方法研究  

Research on Short-term Traffic Flow Prediction Method Based on CS Optimized RVM

作　　者：晏雨婵 YAN Yuchan(College of Electrical Engineering,Shaanxi Polythechnic Institute,Xianyang 712000,China)

机构地区：[1]陕西工业职业技术学院电气工程学院,陕西咸阳712000

出　　处：《微型电脑应用》2024年第12期67-71,共5页Microcomputer Applications

基　　金：咸阳市科技计划重点项目(2019k02-68);陕西工业职业技术学院自然科学类项目(2021YKYB-047)。

摘　　要：智能交通系统中核心内容是短时交通流量预测,因此交通流量预测精度的提高、预测用时的降低成为当前研究的重点问题。针对此问题构建一种布谷鸟搜索(CS)算法优化相关向量机(RVM)回归的短时交通流预测模型。以美国PeMS数据库中的交通流数据为基础,对数据做降噪、归一化处理,以道路占用比和平均速度作RVM预测模型训练集的输入,以交通流量作训练集的输出,采用CS优化算法找到RVM中的核宽度参数σ最优值,提高算法的性能,得到最佳预测模型。通过与RVM、支持向量回归(SVR)预测模型进行比较,所提CS-RVM交通流预测模型的预测精度更高、用时更短。The core content of the intelligent transportation system is short-term traffic flow prediction,so the improvement of traffic flow prediction accuracy and the reduction of prediction time are the key issues of current research.Aimed at this problem,a short-term traffic flow prediction model based on Cuckoo search(CS)algorithm optimized relevance vector machine(RVM)regression is proposed.The traffic flow data are denoised and normalized based on the PeMS.The road occupancy ratio and average speed are used as the input and the traffic flow is used as the output to be the training set in RVM prediction model.It finds the optimal value of the kernel width parameter σ in the RVM by the CS algorithm to improve the performance of the algorithm and obtain the best prediction model.Compared with the RVM and support vector regression(SVR)prediction models,the proposed CS-RVM traffic flow prediction model has higher prediction accuracy and shorter time.

关 键 词：交通流量预测 布谷鸟搜索算法 相关向量机 核宽度参数 

分 类 号：TP181[自动化与计算机技术—控制理论与控制工程]