测控技术2018，Vol.37Issue(2)：33-37,5.
一种平稳化短时交通流预测方法
Stationary Short-Term Traffic Flow Prediction Method
摘要
Abstract
Prediction noise that has the symmetric probability distribution is an assumption condition for the support vector machine (SVM) regression model.However,actual short-term traffic flow data sequence has non-stationary characteristic,which makes it difficult to guarantee the symmetry probability distribution of prediction noise when SVM regression model is used in short term traffic flow prediction,and the prediction precision will deteriorate.In order to solve the above problems,on the basis of proving that the SVM regression model has the symmetry probability distribution for prediction noise of stationary time series,the short-term traffic flow observation sequences with stationary and unstationary are simulated and predicted respectively,and the prediction results are compared and analyzed.The results indicate that the proposed method can reduce the root mean square error of prediction by about 21.6％,reduce the absolute value error by about 21.3％,and reduce the relative error by about 17.3％.The simulation results validate the proposed method.关键词
短时交通流预测/统计学习/平稳化方法/支持向量机/季节性差分Key words
short term traffic flow prediction/statistical learning/stationary method/support vector machine/seasonal difference分类
信息技术与安全科学引用本文复制引用
康军,段宗涛,唐蕾,温兴超..一种平稳化短时交通流预测方法[J].测控技术,2018,37(2):33-37,5.基金项目
国家自然科学基金资助项目(61303041) （61303041）
交通运输部应用基础研究项目(2014319812150) （2014319812150）
陕西省科技厅工业科技攻关项目(2014K05-28,2015GY002,2016GY-078) （2014K05-28,2015GY002,2016GY-078）
