Impact of Real-time Data Filling Methods on Multi-source Data Fusion
-
摘要: 针对各种检测方式中因检测设备失效或因错误数据的排除处理等导致的数据缺失情况,本文提出了时间序列法、空间序列法和历史数据法等3种缺失数据实时补缺方法,并以城市快速路的浮动车数据和微波数据的数据融合为例,通过比较上述方法的补缺精度以及对数据融合精度的影响,分析了不同补缺方法的适用性以及数据补缺处理中的使用优先级.结果表明,基于时间序列和空间序列补缺方法的数据融合结果的平均相对误差均能控制在20%以内,所提出的实时数据补缺方法具有良好的实用性.Abstract: The authers proposed three kinds of real-time data filling methods including time series,spatial correlation and history database methods to complement the missing data,which resulted from detection equipment failure and error data-elimination process.The data filling results based on those methods were applied respectively to an urban expressway multi-source data fusion model,in which,floating car data and remote traffic microwave sensor(RTMS) data were used,and the impact on data-fusion accuracy and the application priority of those data filling methods were analyzed.Resultsshow that the mean absolute percentage errors(MAPEs) of data fusion models based on time series and spatial correlation methods are both under 20%,and that the practicability of the proposed methods is verified.
-
Keywords:
- multi-source data fusion /
- real-time data filling /
- BP neural network
-
-
[1] CHEN C,PETTY K,Skabardonis A,et al.Freewayperformance measurement system:Mining loop detector data[C]∥Annual Meeting of the Transportation ResearchBoard.Washington:Transportation Research Board,2001:96-102.
[2] 姚智胜.基于实时数据的道路网短时交通流预测理论与方法研究[D].北京:北京交通大学交通运输学院,2007.YAO Zhi-sheng.Research on theories and methods ofshort-term traffic flow forecasting of road network based onreal-time[D].Beijing:School of Traffic andTransportation,Beijing Jiaotong University,2007.(inChinese) [3] 姜桂艳,冮龙晖,张晓东,等.动态交通数据故障识别与修复方法[J].交通工程运输学报,2004,4(1):121-125.JIANG Gui-yan,GANG Long-hui,ZHANG Xiao-dong,etal.Malfunction identifying and modifying of dynamic trafficdata[J].Journal of Traffic and TransportationEngineering,2004,4(1):121-125.(in Chinese) [4] WEN Y H,LEE T T,CHO H J.Hybrid models towardtraffic detector data treatment and data fusion[C]∥Proceeding of IEEE International Conference onNetworking,Sensing and Control.Anzona:IEEE Press,2005:525-530.
[5] CHEN C,KWON J,RICE J,et al.Detecting errors andimputing missing data for single loop surveillance systems[C]∥Annual Meeting of the Transportation ResearchBoard.Washington:Transportation Research Board,2003:120-144.
[6] 韦达利,陈锋,卞凯,等.基于数据融合的交通流量数据在线校正[J].中南大学学报:自然科学版,2009,40(1):341-346.WEI Da-li,CHEN Feng,BIAN Kai,et al.On-linecalibration of traffic flow data based on data fusion[J].Journal of Central South University:Science andTechnology,2009,40(1):341-346.(in Chinese) [7] 董长虹.Matlab神经网络与应用[M].2版.北京:国防工业出版社. [8] 张旭.面向交通运行状态评价的多源异质交通流数据融合技术方法研究[D].北京:北京交通大学交通运输学院,2008.ZHANG Xu.Fusion of multi-source heterogeneous trafficflow data for the assessment of traffic operational conditions[D].Beijing:School of Traffic and Transportation,Beijing Jiaotong University,2008.(in Chinese)
计量
- 文章访问数: 13
- HTML全文浏览量: 1
- PDF下载量: 9
下载:
