摘要
为了提高短时交通流预测的精度,提出了基于奇异谱分析和组合核函数最小二乘支持向量机(CKF-LSSVM)的短时交通流预测模型。首先,采用奇异谱分析方法,滤除交通流序列的噪声成分。然后,使用降噪后的交通流数据训练CKF-LSSVM,并通过粒子群优化算法确定模型参数。最后,以厦门市的实测数据为基础,对预测模型进行实验验证和对比分析。结果表明:本文所构建模型具有较好的预测效果,能够有效提高短时交通流预测精度。
In order to improve the accuracy of short-time traffic flow prediction, a short-term traffic flow prediction model based on singular spectrum analysis and Combined Kernel Function （CKF） Least Square Support Vector Machine （LSSVM） is proposed. Singular spectrum analysis technology is used to filter out the noise of traffic time series. Then, the processed traffic flow data are used to train the CKF-LSSVM, and the parameters of the model are determined by particle swarm optimization algorithm. Finally, validation and comparative analysis of the model are carried out using the measured in Xiamen, China. Experimental results indicate that the proposed model has good prediction performance and can effectively improve the accuracy of short-time traffic flow prediction.
出处
《吉林大学学报（工学版）》
EI
CAS
CSCD
北大核心
2016年第6期1792-1798,共7页
Journal of Jilin University:Engineering and Technology Edition
基金
国家科技支撑计划项目(2014BAG03B03)
国家自然科学基金项目(51308248
51408257
51308249)
山东省省管企业科技创新项目(20122150251-1)
关键词
交通运输系统工程
短时交通流预测
奇异谱分析
支持向量机
组合核函数
engineering of communication and transportation system
short-term traffic flow prediction
singular spectrum analysis
support vector machine
combined kernel function
