短时交通流预测模型

樊娜; 赵祥模; 戴明; 安毅生

doi:10.19818/j.cnki.1671-1637.2012.04.015

短时交通流预测模型

doi: 10.19818/j.cnki.1671-1637.2012.04.015

樊娜^1,,
赵祥模¹,
戴明^{1, 2},
安毅生¹

1.
长安大学信息工程学院，陕西西安 710064
2.
中国交通通信信息中心，北京 100011

基金项目:

国家自然科学基金项目 50978030

长江学者和创新团队发展计划项目 IRT0951

陕西省自然科学基金项目 2009-jm8002-1

中央高校基本科研业务费专项资金项目 CHD2011JC027

中央高校基本科研业务费专项资金项目 CHD2011JC056

详细信息

作者简介:
樊娜(1978-), 女, 陕西渭南人, 长安大学讲师, 工学博士, 从事交通信息控制研究

中图分类号: U491.14
计量
- 文章访问数: 1104
- HTML全文浏览量: 158
- PDF下载量: 1077
- 被引次数: 0
出版历程
- 收稿日期: 2012-02-07
- 刊出日期: 2012-08-25

Short-term traffic flow prediction model

1.
School of Information Engineering, Chang'an University, Xi'an 710064, Shaanxi, China
2.
China Transportation Telecommunication and Information Center, Beijing 100011, China

More Information

Author Bio:
FAN Na (1978-), female, lecturer, PhD, +86-29-82334763, fnsea@163.com

摘要

摘要: 针对短时交通流变化周期性与随机性的特点, 提出了新的混合预测模型, 包含非参数回归模型与BP神经网络模型2种单项模型。非参数回归模型利用相关历史交通流数据, 通过数据库匹配操作, 确定预测结果, 以充分体现交通流的周期稳定性。采用3层BP神经网络模型反映交通流的动态与非线性特点。采用模糊控制算法确定各单项模型的权重, 并按不同权重有效组合成新的混合模型。采用西安市某路段30 d的交通流量数据验证混合模型的预测效果。试验结果表明: 该混合模型的平均相对误差为1.26%, 最大相对误差为3.53%, 其预测精度明显高于单项模型单独预测时的精度, 能较准确地反映交通流真实情况。
- 短时交通流预测 /
- 混合模型 /
- 非参数回归 /
- BP神经网络 /
- 模糊控制
Abstract: A new hybrid prediction model including two single models of nonparametric regression model and BP neural network model was proposed according to the periodicity and randomness properties of short-term traffic flow.Relevant historical traffic flow data were used in nonparametric regression model to make the prediction result abtained from the databases matching proceeding fully illustrate the cyclical stability of traffic flow.Three-tier BP neural network model was used to reflect the dynamic and nonlinear characters of traffic flow.Fuzzy control algorithm was adopted to get the weight coefficient of each model.New mixed model was constituted by the two single models according to different weight coefficients.The prediction effect of hybrid prediction model was verified by the traffic flow data in 30 d from a certain section in Xi'an.Experimental result indicates that the average relative error of mixed model is 1.26%, and its maximum relative error is 3.53%, so the prediction accuracy of mixed model is obviously higher than two single models, and can accurately reflect the real situation of traffic flow.
- short-term traffic flow perdiction /
- hybrid model /
- nonparametric regression /
- BP neural network /
- fuzzy control

HTML全文

图 1 非参数回归模型预测流程

Figure 1. Prediction procedure of nonparametric regression model

下载: 全尺寸图片幻灯片

图 2 预测路段与上游路段交通流量的关系

Figure 2. Relationship of traffic flows between prediction section and upstream sections

下载: 全尺寸图片幻灯片

图 3 试验路段

Figure 3. Test sections

下载: 全尺寸图片幻灯片

图 4 三种模型预测结果对比

Figure 4. Comparison of prediction result in 3 models

下载: 全尺寸图片幻灯片

图 5 预测结果与实测数据对比

Figure 5. Comparison between prediction result and real data

下载: 全尺寸图片幻灯片

表 1 模糊变量的定义

Table 1. Definitions of fuzzy variables

下载: 导出CSV

表 2 部分模糊规则定义

Table 2. Definitions of partial fuzzy rules

下载: 导出CSV

表 3 中间层神经元个数

Table 3. Neuron numbers of middle layer

下载: 导出CSV

表 4 输入量的预测结果

Table 4. Prediction result of inputs

下载: 导出CSV

表 5 评价指标对比

Table 5. Comparison of evaluation indexes

下载: 导出CSV

表 6 不同数据库规模下3种方法的预测结果

Table 6. Prediction results of three methods under different database sizes

下载: 导出CSV

参考文献(16)

[1]	VLAHOGIANNI E I, KARLAFTIS M G, GOLIAS J C. Optimized and meta-optimized neural networks for short-term traffic flow prediction: agenetic approach[J]. Transportation Research Part C: Emerging Technologies, 2005, 13(3): 211-234. doi: 10.1016/j.trc.2005.04.007
[2]	王进. 短期交通流预测模型与方法研究[D]. 北京: 清华大学, 2005. WANG Jin. Studies on short-term traffic flow forecasting models and methods[D]. Beijing: Tsinghua University, 2005. (in Chinese).
[3]	戴施华, 周欣荣. Kalman滤波理论在短时交通预测上的应用[J]. 哈尔滨商业大学学报: 自然科学版, 2005, 21(6): 728-730, 735. doi: 10.3969/j.issn.1672-0946.2005.06.014 DAI Shi-hua, ZHOU Xin-rong. Application of Kalman filtering theory on prediction of short-term traffic volume[J]. Journal of Harbin University of Commerce: Natural Sciences Edition, 2005, 21(6): 728-730, 735. (in Chinese). doi: 10.3969/j.issn.1672-0946.2005.06.014
[4]	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, 16(1): 82-86. GONG Xiao-yan, TANG Shu-ming. Integrated traffic flow forecasting and traffic incident detection algorithm based on non-parametric regression[J]. China Journal of Highway and Transport, 2003, 16(1): 82-86. (in Chinese).
[5]	高慧, 赵建玉, 贾磊. 短时交通流预测方法综述[J]. 济南大学学报: 自然科学版, 2008, 22(1): 88-94. doi: 10.3969/j.issn.1671-3559.2008.01.024 GAO Hui, ZHAO Jian-yu, JIA Lei. Summary of short-time traffic flow forecasting methods[J]. Journal of University of Jinan: Science and Technology, 2008, 22(1): 88-94. (in Chinese). doi: 10.3969/j.issn.1671-3559.2008.01.024
[6]	赵建玉, 贾磊, 杨立才, 等. 基于粒子群优化的RBF神经网络交通流预测[J]. 公路交通科技, 2006, 23(7): 116-119. doi: 10.3969/j.issn.1002-0268.2006.07.028 ZHAO Jian-yu, JIA Lei, YANG Li-cai, et al. RBF neural network traffic flow forecasting model based on particle swarm optimization[J]. Journal of Highway and Transportation Research and Development, 2006, 23(7): 116-119. (in Chinese). doi: 10.3969/j.issn.1002-0268.2006.07.028
[7]	况爱武, 黄中祥. 基于RBF神经网络的短时交通流预测[J]. 系统工程, 2004, 22(2): 63-65. https://www.cnki.com.cn/Article/CJFDTOTAL-GCXT200402012.htm KUANG Ai-wu, HUANG Zhong-xiang. Short-term traffic flow prediction based on RBF neural network[J]. Systems Engineering, 2004, 22(2): 63-65. (in Chinese). https://www.cnki.com.cn/Article/CJFDTOTAL-GCXT200402012.htm
[8]	MAI T, GHOSH B, WILSON S. Multivariate short-term traffic flow forecasting using Bayesian vector autoregressive moving average model[C]∥TRB. Transportation Research Board91 st Annual Meeting. Washington DC: TRB, 2012: 3728-3740.
[9]	CASTRO-NETO M, JEONG Y S, JEONG M K, et al. Online-SVR for short-term traffic flow prediction under typical and atypical traffic conditions[J]. Expert Systems with Applications, 2009, 36(3): 6164-6173. doi: 10.1016/j.eswa.2008.07.069
[10]	ZHANG Yang, LIU Yun-cai. Application of combined fore-casting models to intelligent transportation systems[J]. Studies in Computational Intelligence, 2009, 214(10): 181-186. doi: 10.1007/978-3-540-92814-0_28
[11]	李元诚, 李波, 方廷健. 基于小波支持向量机的非线性组合预测方法研究[J]. 信息与控制, 2004, 33(3): 303-306, 324. doi: 10.3969/j.issn.1002-0411.2004.03.011 LI Yuan-cheng, LI Bo, FANG Ting-jian. Nonlinear combin-ation forecasting method based on wavelet support vector machines[J]. Information and Control, 2004, 33(3): 303-306, 324. (in Chinese). doi: 10.3969/j.issn.1002-0411.2004.03.011
[12]	谭满春, 冯荦斌, 徐建闽. 基于ARIMA与人工神经网络组合模型的交通流预测[J]. 中国公路学报, 2007, 20(4): 118-121. https://www.cnki.com.cn/Article/CJFDTOTAL-ZGGL200704022.htm TAN Man-chun, FENG Luo-bin, XU Jian-min. Traffic flow prediction based on hybrid ARIMA and ANN model[J]. China Journal of Highway and Transport, 2007, 20(4): 118-121. (in Chinese). https://www.cnki.com.cn/Article/CJFDTOTAL-ZGGL200704022.htm
[13]	DAVIS G A, NIHAN N L. Nonparametric regression and short-term freeway traffic forecasting[J]. Journal of Trans-portation Engineering, 1991, 117(2): 178-188. doi: 10.1061/(ASCE)0733-947X(1991)117:2(178)
[14]	SMITH B L, WILLIAMS B M, OSWALD R K. Comparison of parametric and nonparametric models for traffic flow fore-casting[J]. Transportation Research Part C: Emerging Tech-nologies, 2002, 10(4): 303-321. doi: 10.1016/S0968-090X(02)00009-8
[15]	甘健胜, 陈国龙. 线性组合预测模型及其应用[J]. 计算机科学, 2006, 33(9): 191-194. doi: 10.3969/j.issn.1002-137X.2006.09.058 GAN Jian-sheng, CHEN Guo-long. Linear combination fore-casting model and its application[J]. Computer Science, 2006, 33(9): 191-194. (in Chinese). doi: 10.3969/j.issn.1002-137X.2006.09.058
[16]	饶从军, 王成, 涂火年. 一种新的预测模型及其应用[J]. 武汉理工大学学报: 交通科学与工程版, 2007, 31(6): 1098-1101. doi: 10.3963/j.issn.2095-3844.2007.06.041 RAO Cong-jun, WANG Cheng, TU Huo-nian. Novel fore-casting model and its application[J]. Journal of Wuhan University of Technology: Transportation Science and Engin-eering, 2007, 31(6): 1098-1101. (in Chinese). doi: 10.3963/j.issn.2095-3844.2007.06.041