基于路网相关性的分布式增量交通流大数据预测方法

doi:10.13249/j.cnki.sgs.2017.02.006

摘要/Abstract

摘要：

针对城市道路拥堵问题的日益加剧的问题,智能化城市交通管理平台是缓解拥堵问题的有效方法,利用交通流大数据预测结果进行交通诱导,能够指导用户调整出行方案,有效缓解交通压力。研究了交通流大数据的分布式增量聚合方法,对海量交通流数据进行清洗统计,为交通流预测提供数据基础,基于交通流在路网中上下游路段的相关性分析,利用路口转弯率多阶分配将该相关性量化,构建基于路网相关性的空间权重矩阵,完成对于STARIMA模型的改进。通过应用试验证明,该方法能更准确的进行交通流预测,为交通诱导信息发布提供依据。

关键词: 交通流, 大数据, 分布式增量, 路网相关性, STARIMA

Abstract:

Along with the accelerating urbanization, there are more and more contradictions between the number of cars and urban transportation facilities. The congestion time and congested roads in cities are increasing. Intelligent urban traffic management platform is the effective method to alleviate the increasingly serious urban congestion problems. By using prediction results of traffic flow big data, the platform can guide users to adjust the travel plan, and ease the traffic pressure effectively. How to use a large number of spatio-temporal data related to traffic activities to predict the traffic flow is the key to realizing traffic guidance. In this article, a distributed incremental aggregation method for traffic flow data is studied. The method combines the distributed incremental data aggregation method with the traffic flow data cleaning rules, makes cleaning and counting of traffic flow big data, and provides data for traffic flow forecast. With the analysis of traffic flow correlation in the network of upstream and downstream, this article uses the multi-order of turning rate in the intersection to quantize the correlation, builds the spatial weight matrix based on the road network correlation, and improves the STARIMA model. In this article, two groups of contrast experiments were made. Through the contrast experiment between MapReduce method and MPI method, the result proves that the method proposed in this article is better than the MPI method in the development cycle and stable operation. The method’s efficiency can meet the need of traffic flow data aggregation. The traffic flow statistics can be used as the basis of traffic flow forecasting. Through the contrast experiment between the Improved STARIMA model and the Dynamic STARIMA model, the result proves that the Improved STARIMA model, which considers the multi-order correlation between the upstream and downstream sections, matches the distribution rules of traffic flow in road network better. Therefore, the forecast results are more accurate. In conclusion, the method of this article is a new method of traffic flow forecasting in the background of big data, and it can realize accurate prediction.

Key words: traffic flow, big data, distributed incremental, road network correlation, STARIMA

中图分类号:

K909

李欣, 孟德友. 基于路网相关性的分布式增量交通流大数据预测方法[J]. 地理科学, 2017, 37(2): 209-216.

Xin Li, Deyou Meng. Distributed Incremental Traffic Flow Big Data Forecasting Method Based on Road Network Correlation[J]. SCIENTIA GEOGRAPHICA SINICA, 2017, 37(2): 209-216.

图/表 6

图1

图2

图3

图4

图5

表1

参考文献 30

[1]	李德仁,马军,邵振峰.论时空大数据及其应用[J].卫星应用,2015,(9):7-11.
	[Li Deren, Ma Jun, Shao Zhenfeng.The theory of space-time big data and its application. Satellite Application, 2015, (9):7-11.]
[2]	Bose J H, Andrzejak A, Hogqvist M.Beyond online aggregation: Parallel and incremental data mining with online Map-Reduce[M]//Proc of Workshop on Massive Data Analytics on the Cloud. New York:ACM,2010
[3]	Aghabozorgi S, Saybani M R, Wah T Y.Incremental clustering of time-series by fuzzy clustering. Journal of Information Science and Engineering,2012,28(4):671-688
[4]	Laptev N, Zeng K, Zaniolo C.Very fast estimation for result and accuracy of big data analysis:The EARL system[M]//Proc of ICDE.Piscataway,NJ:IEEE,2013:1296-1299.
[5]	Zhang S B, Han J Z,Liu Z Y et al. Accelerating MapReduce with Distributed Memory Cache[M]//Proc of ICPADS. Piscataway, NJ:IEEE, 2009:472-478.
[6]	Stephanedes Y J, Michalopoulos P G.Improved estimation of traffic flow for real-time control[M]// Transportation Research Record 795, Washington DC: Transportation Research Board, 1981:28-39.
[7]	Okutani I, Stephanedes Y J.Dynamic prediction of traffic volume through Kalman filtering theory[J]. Transportation Research Part B:Methodological, 1984,18(1):1-11. doi: 10.1016/0191-2615(84)90002-X
[8]	Ahmaed M S, Cook A R.Analysis of freeway traffic time-series data by using Box-Jenkins technique[M]//Transportation Research Record 722. Washington DC: Transportation Research Board, l 979:l-9.
[9]	Doughetry M S, Cobbett M R.Short-term inter-urban traffic forecasts using neural networks[J]. International Journal of Forecasting, l997, 13(1):2l-31. doi: 10.1016/S0169-2070(96)00697-8
[10]	Ledoux C.An urban traffic flow model integrating neural networks[J]. Transportation Research Part C: Emerging Technologies, 1997, 5(5):287-300. doi: 10.1016/S0968-090X(97)00015-6
[11]	Yue Yang.Spatial-temporal dependency of traffic flow and its implications for short-term traffic forecasting[D]. Hong Kong: The University of Hong Kong, 2006.
[12]	Kamarianakis Y, Prastacos P.Space-time modeling of traffic flow[J]. Computers & Geosciences, 2005, 31:119-133. doi: 10.1016/j.cageo.2004.05.012
[13]	Martin R L, Oeppen J E.The identification of regional forecasting models using space-time correlation functions[J]. Trans Inst Brit Geogr, 1975, 66: 95-118. doi: 10.2307/621623
[14]	余碧莹,邵春福.基于时空模型的道路网交通状态预测[M]//第四届中国智能交通年会论文集.青岛:全国智能交通系统协调指导小组和山东省人民政府, 2008: 546-551.
	[Yu Biying, Shao Chunfu.Traffic state forecast of road network based on space-time model//Proceedings of the Fourth China Annual Conference on ITS. Qingdao: The intelligent transportation system to coordinate and guide team and the People’s Government of Shandong Province, 2008:546-551.]
[15]	Lin Shulan, Huang Hongqiang, Zhu Daqi et al. The application of space-time ARIMA model On traffic flow forecasting[M] // Proceedings of the 8th International Conference on Machine Learning and Cybernetics. Baoding: Hebei University and IEEE SMC Association, 2009:3408-3412.
[16]	Min Xinyu, Hu Jianming, Chen Qi et al. Short-term traffic flow forecasting of urban network based on dynamic STARIMA model[M] // Proceedings of the 12th International IEEE Conference on Intelligent Transportation Systems. St. Louis, MO, USA: Institute of Electrical and Electronics Engineers, 2009:461-466.
[17]	瞿莉. 基于动态交通流分配参数的网络交通状态建模与分析[D].北京:清华大学,2010.
	[Qu Li.Modeling and analyzing the network-level traffic status based on dynamic traffic assignment ratios. Beijing: Tsinghua University, 2010.]
[18]	张和生,张毅,胡东成,等.区域交通状态分析的时空分层模型[J].清华大学学报:自然科学版,2007, 47(1):157-160. doi: 10.3321/j.issn:1000-0054.2007.01.042
	[Zhang Hesheng, Zhang Yi, Hu Dongcheng et al. Spatial-temporal hierarchical model for area traffic state analysis. Journal of Tsinghua University: Sci &Technol. 2007, 47(1):157-160.] doi: 10.3321/j.issn:1000-0054.2007.01.042
[19]	王晓原,张敬磊.交通流信息挖掘的非参数概率变点模型研究[J].武汉理工大学学报:交通科学与工程版, 2010, 34(4):801-805. doi: 10.3963/j.issn.1006-2823.2010.04.039
	[Wang Xiaoyuan, Zhang Jinglei.Study on Nonparametric Probability Change-point Model for Traffic Flow Exploitation. Journal of Wuhan University of Technology (Transportation Science & Engineering), 2010, 34(4):801-805.] doi: 10.3963/j.issn.1006-2823.2010.04.039
[20]	郭志懋,周傲英.数据质量和数据清洗研究综述[J].软件学报, 2002, 13(11): 2076-2082.
	[Guo Zhimao, Zhou Aoying.Research on Data Quality and Data Cleaning: A Survey. Journal of Software, 2002, 13(11): 2076-2082.]
[21]	张敬磊,王晓原.基于非线性组合模型的交通流预测方法[J].计算机工程, 2010, 36(5):202-205.
	[Zhang Jinglei, Wang Xiaoyuan.Traffic Flow Prediction Method Based on Non-linear Hybrid Model[J]. Computer Engineering, 2010, 36(5):202-205.]
[22]	王晓原,张敬磊,吴芳.交通流数据清洗规则研究[J].计算机工程, 2011, 37(20):191-193. doi: 10.3969/j.issn.1000-3428.2011.20.066
	[Wang Xiaoyuan, Zhang Jinglei, Wu Fang.Research on Traffic Flow Data Cleaning Rules. Computer Engineering, 2011, 37(20):191-193.] doi: 10.3969/j.issn.1000-3428.2011.20.066
[23]	Pfeifer P E, Deutsch S J.A three-stage iterative procedure for space-time modeling[J]. Technometrics, 1980, 22(1): 35-47. doi: 10.1080/00401706.1980.10486099
[24]	Pfeifer P E, Deutsch S J.Identification and interpretation of first-order space-time ARMA models[J]. Technometrics, 1980, 22(3):397-408. doi: 10.1080/00401706.1980.10486172
[25]	Pfeifer P E, Deutsch S J.Variance of the sample-time autocorrelation function of contemporaneously correlated variables[J]. SIAM Journal of Applied Mathematics, Series A, 1981, 40(1):133-136. doi: 10.1137/0140010
[26]	Deng Shuo, Hu Jianming, Wang Yin et al. Urban road network modeling and real-time prediction based on house holder transformation and adjacent vector[M] //Advances in Neural Networks—ISNN 2009. Berlin Heidelberg: Springer, 2009:899-908.
[27]	Bezdek J C, Pal N R.Some new indexes of cluster validity[J]. IEEE Trans Syst Man Cy, 1998, 28: 301-315.
[28]	Kamarianakis Y, Prastacos P.Space-time modeling of traffic flow[J]. Comput Geosci-UK, 2005, 31: 119-133. doi: 10.1016/j.cageo.2004.05.012
[29]	牛新征,佘堃.面向大规模数据的快速并行聚类划分算法研究[J].计算机科学,2012,(1):134-137,151. doi: 10.3969/j.issn.1002-137X.2012.01.030
	[Niu Xinzheng, She Kun.Study of Fast Parallel Clustering Partition Algorithm for Large Data Sets.Computer Science, 2012,(1):134-137,151.] doi: 10.3969/j.issn.1002-137X.2012.01.030
[30]	Smith B L, Demeisky M J.Traffic flow forecasting: comparison of modeling approaches[J]. Journal of Transportation Engineering, 1997,123(4):261-266.

编号	动态 STARIMA	路网相关STARIMA	编号	动态 STARIMA	路网相关STARIMA	编号	动态 STARIMA	路网相关STARIMA
1	8638.27	6643.68	22	3338.29	2839.48	43	3977.67	3058.83
2	7396.01	6032.39	23	6236.77	5064.66	44	4458.98	3847.45
3	6631.31	4983.57	24	7784.39	5338.54	45	8979.37	7432.82
4	8362.58	5986.44	25	2526.67	1438.33	46	8443.69	6234.73
5	2948.21	1863.49	26	3464.32	1974.34	47	2798.55	1846.49
6	4820.22	2799.35	27	8764.39	5890.43	48	3985.42	2275.45
7	9230.23	7390.32	28	6549.48	5438.93	49	7849.43	5893.37
8	7857.46	6074.45	29	12974.45	8865.29	50	6692.38	5624.78
9	15324.01	12984.61	30	13084.44	10474.63	51	6639.32	4542.43
10	11479.46	8753.05	31	5478.34	3740.35	52	5873.57	4147.57
11	3892.27	2775.73	32	4858.34	3275.39	53	3720.32	2475.54
12	4917.48	3295.48	33	7434.95	6578.36	54	3920.28	2143.27
13	3729.33	2903.57	34	7868.23	5920.49	55	2039.58	1343.47
14	3928.23	2638.57	35	6743.23	4839.33	56	3235.62	1634.57
15	4478.19	3902.29	36	7820.33	5923.67	57	4838.88	3822.44
16	6903.36	4632.84	37	8068.35	6488.38	58	5749.23	4727.28
17	10573.48	7296.23	38	7819.28	6367.35	59	6403.45	4884.74
18	9033.54	7018.83	39	3894.22	2057.75	60	6653.63	4954.54
19	4847.34	3892.33	40	4780.44	3628.65	61	7570.49	6343.45
20	3309.21	3087.88	41	7897.36	4929.38	62	8788.48	6932.43
21	5789.22	4274.49	42	8902.34	5563.37