摘要
获取有关交通流的实时信息是实现智能交通的重要步骤之一。在智能交通系统中,道路上的动态交通流预测是交通监控系统的基本组成部分。拥堵管理、车辆动态路线引导、公交管理、出行者信息系统、突发事件检测系统等都需要参考指定路段交通流状况。一种好的交通流量预测方法可以为交通控制提供真实可靠的数据信息。一般的交通流预测分为短期预测和长期预测,短期预测具有实时性,它很容易适应即时数据发生的改变,长期预测主要体现交通流在一个较长时段内的规律性,并不是所有状态改变都是永久的。本文基于已有的预测方法,提出一种短期、长期组合的预测方法,以交通流量预测为依托,利用从交通监控系统中得到的数据信息,在具体路段进行预测方法的验证。
本文先利用双指数平滑进行短期预测,在双指数平滑预测中利用Levenberg-Marquardt算法优化平滑参数,以减小人工选择参数的误差；然后利用傅立叶级数对前期预测序列和实际观测序列之间的残差进行修正；最后基于马尔可夫状态转移模型在长期进行调整。本文将长、短期策略综合考虑提出一种组合的方法,对交通流数据进行预测并修正,使得预测结果既能很好的反映实时性,又能体现准确性和适应性。
实例验证时,基于交通流数据具有随机性、自组织、类周期性等特征,在执行预测前对实际测量数据进行预处理去噪平滑。实验证明该模型既利用实时数据提供的当前方式变化特征信息,也能把握历史数据所提供的方式变化总体规律信息。模型验证对比结果表明,本文提出的模型有较高的预测精度。
Getting the real-time information of traffic flows was one of the important steps to achieve the intelligent transportation. In the Intelligent Transportation Systems, the forecasting of dynamic traffic flow on the road was an essential part of the Traffic Monitoring System. The traffic flow conditions of specified sections were the necessary reference for congestion management, vehicle's dynamic route guidance, traffic management, traveler information system and incident detection system. A good traffic flow forecasting methods could provide true and reliable data for traffic control. Prediction of traffic flow was divided into short-term forecasting and long-term forecasting. The short-term forecasting had the real-time which made it adapt to the change of the real-time data easily. The long-term forecasting mainly reflected the regularity of traffic flow in a longer period of time, for the reason that not all state changes were permanent. This paper put forward the methods combinated the short-term and long-term prediction based on the existing prediction methods. It relied on the traffic projections, and used the data obtained from the Traffic Monitoring System to make validation of prediction methods.
Firstly, this article used the double exponential smoothing for the short-term forecasting. In the double exponential smoothing forecasting, it used Levenberg-Marquardt algorithm to optimize the smoothing parameters so that the error of artificial selection parameters could be reduced. Then the residuals between the pre-prediction sequence and observation sequence could be corrected by the applying of the Fourier series. Last, after long-term adjustment based on the Markov state transition model, it proposed a combination method that could predict and correct the traffic flow data. This made the forecasted results not only reflecting real-time, but also reflecting the accuracy and adaptability.
In the process of the instance validation, because of the characteristics of the traffic flow data such as randomness, self-organization and class cyclical, we must make denoising smooth for preprocessing over the actual measurement data before the implementation of forecasting. Experiments show that this model used both information of the current way to change the characteristics provided by real-time data and the overall law information of the way to change provided by historical data. The comparison result of model validation shows that the proposed model of the paper has better prediction accuracy.
引文
[1]王正武,黄中祥.短时交通流预测模型的分析与评价[J].系统工程,2003,6(21)：97-100.
[2]刘静,关伟.交通流预测方法综述[J].公路交通科技,2004,21：82-85.
[3]S. Makridakis and S.C. Wheelwright:Forecasting Methods for Management, John Wiley & Sons,1989:13-16.
[4]王科伟.高速公路服务水平评价研究[J].中国公共安全：智能交通,2006(5).
[5]Filardo, A.J.,(1994).Business-cycle phase and their transitional dynamics. Jouranal of Business and Economic Statistics,12,299-308.
[6]谭国贤,翁小熊,姚树中等.城市交义口的短时交通流预测模型[J].交通与计算科技,2003.8.
[7]乔德华,张开禾,范耀祖.多模型交通流预测优化[J].交通标准化,2007(4).
[8]朱中,杨兆升.实时交通流量人工神经网络预测模型[J].中国公路学报,1998(4)：89-92.
[9]罗志增,蒋静坪.相关证据的融合及其在机器人多传感器信息融合中的应用.传感技术学报,2000,3：177-182.
[10]苏浩伟.智能交通系统及共用信息平台关键技术的研究及应用[D].广州：华南理工大学,2009.
[11]王雷,许晗,郑山水.ITS中交通流预测方法的研究[J].中国科技信息,2008,10：268-269.
[12]郭敏,肖翔,蓝金辉.道路交通流短时预测方法综述[J].自动化技术与应用,2009,6(28)：8-9
[13]陆海亭,张宁,黄卫,夏井新.短时交通流预测方法研究进展[J].交通运输工程与信息学报,2009,4(7)：84-91.
[14]宫晓燕,汤淑明.基于非参数回归的短时交通流量预测与事件检测综合算法[J].中国公路学报,2003,(1)：82-86.
[15]陈玫.北京市公共交通服务需求量双指数平滑法预测分析[J].武汉理工大学学报(交通科学与工程版),2008,4(32)：661-664.
[16]王冬星,朱建秋,杨引霞.一种指数平滑预测的参数优化方法及实现[J].微机发 展,2005,3(15):1-3.
[17]吴怀宇.时间序列分析与综合[M].武汉大学出版社,2004.
[18]Joseph J.,LaViola Jr. An Experiment Comparing Double Exponential Smoothing and Kalman Filter-Based Predictive Tracking Algorithms[J]. Proceedings of virtual reality, 2003:283-284.
[19]杨云霞.时间序列预测模型及其应用[J].太原师范学院学报,2005,4(4):4-7.
[20]Vlahogianni E.I., Karlaftis M.G. and Golias J C. Statistical methods for detecting non-linearity and non-stationarity in univariate short-term time-series of traffic volume[J]. Transportation Research Part C,2006,14:351-367.
[21]翟亮亮,王连堂,王俊杰.基于Levenberg-Marquardt算法的油田产量预测模型[J].2010,3(24):53-55.
[22]胡海琴,蒋传文,蔡宏欣.中长期负荷预测的傅里叶级数残差修正模型[J].2010,2(27):39-41
[23]甘丽珍,张铭鑫,刘明周,蒋增强,葛茂根.基于傅里叶变换的质量预测动态残差修正模型[J].合肥工业大学学报(自然科学版),2009,6(32):806-810.
[24]魏巍贤,陈智文,王建军.三状态马尔可夫机制转换模型研究—在世界油价波动分析中的应用[J].财经研究,2006,6.
[25]孙湘海,刘潭秋.基于马尔柯夫制度转换模型的短期交通流预测[C]//2008年国际交通技术创新与应用大会暨国际交通基础设施建设与养护技术大会论文集.北京：交通部科学研究院,2008.
[26]陈科杰,夏洪胜.应用统计在经济管理中的应用—论马尔柯夫链与市场占有率的预测与分析[J].商场现代化,2004(10).
[27]王建军Markov机制转换模型研究—在中国宏观经济周期分析中的应用[J].数量经济技术经济研究.2007(3).
[28]Hamilton, James D. A New Approach to the Economic Analysis of Nonstationary Time Series and the Business Cycle. Econometrica,1989,57:357-384.
[29]Hamilton, J.D.(1994). Time series analysis, Princeton, NJ:Princeton University Press
[30]Rence Garcia,Pieerre Perron. Ananalysis of the real interest rate under regime shifts. The Review of Economics and Statistics,1996,78(1):111-125.
[31]Smith PA.How well do Markov Switching Models Describe Business Cycles, Journal of Applied Econometric,2005,20(1):253-74.
[32]T C Chm,J Ang,N S Seah,D P Mital and H C Chua, Markov-Switching Regime for Time Series Prediction, Fourth International Conference on knowledge-Based,2000, 8(2):470-744.
[33]谢赤,刘潭秋.人民币实际汇率中的马尔可夫转换行为[J].统计研究,2003,9:50-52.
[34]贺国光,徐岩宇.车辆线路引导系统的行驶时间预测模型研究[J].公路学报,1998,11(3)：18-21.
[35]Dai shihua,Zhou xinrong. Application of kalman filtering theory on prediction of short-term traffic volume[J].Journal of Harbin Commercial University.2005(6):728-730.
[36]Wangjin. Studies on short-term traffic flow forecasting models and methods[D]. Beijing:Tsinghua University.2005.
[37]Weigend A S, Rumelhart D E and Huberman B A.Generalization by weight-elimination applied to currency exchange rate prediction. Proc. Int. Joint Conf.Neural Networks. Seattle, WA.1991,1:837-841
[38]沙云飞等.基于GPS的路段旅行时间和速度估算算法研究[C].北京：全国ITS年会论文集,2006.
[39]Nam Do.H.and Drew Donald R.Method for Estlination Freeway Travel Tim in Real Time from Flow Measurements[J].Journal of Transportation Engineering,1996:185-191.
[40]Huang, h.-q, t.-h. Tang. Short-term traffic flow forecasting based on ARIMA-ANN[C] //Control and Automation.ICCA 2007.IEEE International Conference on.2007.
[41]孙立光,李瑞敏,董屾,陆化普.短时交通流组合预测模型研究[J].武汉理工大学学报(交通科学与工程版),2010,5(35):874-876.
[42]唐明.短时交通流特性及其预测方法的研究[D].湖南：长沙理工大学,2004.
[43]良建俊.车辆检测器及道路流量检测系统的研究设计[J].科技情报开发与经济,2009,(35).
[44]王珺,吴兵,王靖阳.ITS实时检测交通数据的质量控制技术探讨[J].交通与运 输,2005.(4).