欢迎来到天天文库
浏览记录
ID:34801951
大小:5.04 MB
页数:69页
时间:2019-03-11
《基于小波理论的短时交通流预测方法研究》由会员上传分享,免费在线阅读,更多相关内容在学术论文-天天文库。
1、长沙理工大学硕士学位论文基于小波理论的短时交通流预测方法研究姓名:喻丹申请学位级别:硕士专业:交通运输规划与管理指导教师:吴义虎20080320摘要实时、准确的短时交通流预测是智能交通控制与管理的基础,许多预测方法被提出,但是,因未考虑短时交通流中不确定干扰因素的影响,或者将各干扰简单复合统一处理,预测结果准确性较差。本文从短时交通流特性分析出发,利用PCA主成分分析法和分形理论验证短时交通流是一组包含干扰信号的混沌序列,具有最小可预测周期。短时交通流成分复杂,不同特性信号成分在预测中的作用不同,因此本文基于改进的Mallat算法进行小波分解和单支重构,将短时交通流分离成低频
2、确定信号、高频混沌信号和高频干扰信号。对各分解信号,构造双层小波网络分别预测:第一层小波网络WNN.1用于低频确定信号和高频干扰信号的预测;第二层小波网络WNN.2用于高频混沌信号的预测。最后,将各分解信号预测值迭加以获得包含原始信号所有特性成分的预测值。算例研究表明,本文提出的双层小波网络短时交通流预测法具有较高预测精度和较快的预测速度。关键词:短时交通流;双层小波网络;多分辨;信号分解;混沌AbstractThecontrollingandmanagementofintectlencetrafficisbasedonthereal-timeandaccurateforec
3、astingofshort·termtrafficflow.Agreatmanyofforecastingapproachesforshort—termtrafficflowareadvanced.Buttheprecisionoftheseforecastingapproachesaredissatisfiedduetotheinfluenceofalltheinterferencedon’tbeconsideredfullyorthecombinedactionofinterferenceistransactedbysinglemethod.Sothecharacteri
4、sticofshort—termtrafficflowisanalyzedfirstbymethodnamedPrincipalComponentAnalysisandFractalTheoryinthepaper.Wecouldconcludethattheshort—termtrafficflowpossessedminimalpredictablecycleisasetofchaotictimeserialcontainingmuchinterference,andthedifferentcomponentinshort·termtrafficflowowneddiff
5、erentcharacteristichasdifferentinfluencefortheforecastingresults.SotheimprovedMallatalgorithmisadvancedinthepapertocompleteinformationseparation,whichisamethodusedforshort—termtrafficflowwaveletdecompositionandsingle-branchreconstruction,technically.Andtheshort-termtrafficflowtimeserialrese
6、archedinthepapercouldbeseparatedintolow·ffequencydetermineinformation,high·frequencychaoticinformationandhigh-frequencyinterferenceinformation.Thenadouble·layerwaveletnetworkisestablishedtoforecasttheinformationseparated:thefirstlayerofwaveletnetworkismarkedasWNN·1,whichisusedtoforecastthel
7、ow-·frequencydetermineinformationandhigh·-frequencyinterferenceinformation;thesecondlayerofwaveletnetworkiSmarkedasWNN·2,whichisusedtoforecastthehigh-矗equencychaoticinformation.Intheend,wecouldgettheforecastingresultsofshort—termtrafficflowbysuperposingt
此文档下载收益归作者所有