<i>K</i>近邻短时交通流预测模型

于滨; 邬珊华; 王明华; 赵志宏

doi:10.19818/j.cnki.1671-1637.2012.02.015

K近邻短时交通流预测模型

doi: 10.19818/j.cnki.1671-1637.2012.02.015

1.
大连海事大学交通运输管理学院, 辽宁大连 116026
2.
长安大学信息工程学院, 陕西西安 710064

基金项目:

国家自然科学基金项目 51108053

中国博士后科学基金项目 201003611

中央高校基本科研业务费专项资金项目 2011ZC029

中央高校基本科研业务费专项资金项目 2011QN037

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:
于滨(1977-), 男, 辽宁大连人, 大连海事大学副教授, 工学博士, 从事智能交通系统研究

中图分类号: U491.14
计量
- 文章访问数: 944
- HTML全文浏览量: 198
- PDF下载量: 797
- 被引次数: 0
出版历程
- 收稿日期: 2011-10-29
- 刊出日期: 2012-04-25

K-nearest neighbor model of short-term traffic flow forecast

1.
School of Transportation Management, Dalian Maritime University, Dalian 116026, Liaoning, China
2.
School of Information Engineering, Chang'an University, Xi'an 710064, Shaanxi, China

More Information

Author Bio:
YU Bin (1977-), male, associate professor, PhD, +86-411-84726756, yubinyb@163.com

摘要

摘要: 为了准确预测道路短时交通流, 构建了基于K近邻算法的短时交通流预测模型。分析了K近邻算法的时间和空间参数, 提出4种状态向量组合的K近邻模型: 时间维度模型、上游路段-时间维度模型、下游路段-时间维度模型与时空参数模型。以贵州省贵阳市出租车的GPS数据对几种K近邻模型进行了检验。分析结果表明: 带有时空参数的K近邻模型具有更高的预测精度, 其预测误差最小, 平均为7.26%。基于指数权重的距离度量方式能更精确的选择近邻, 其预测误差最小, 平均为5.57%。与神经网络和历史平均模型相比, 带有指数权重的K近邻模型具有更好的预测精度, 平均预测误差仅为9.43%。可见, 带有时空参数与指数权重的K近邻模型可作为道路短时交通流预测的有效手段。
- 交通信息工程 /
- 短时交通流预测 /
- K近邻模型 /
- 时空参数 /
- 指数权重
Abstract: In order to accurately forecast the short-term traffic flow, a K-nearest neighbor(K-NN) model was set up.The time and space parameters of the K-NN model were analyzed.Based on four different combinations of state vectors, the time dimension model, upstream section-time dimension model, downstream section-time dimension model and space-time dimension model were proposed.The four different models were validated by using the GPS data from taxis of Guiyang.Analysis result indicates that the K-NN model with both space and time parameters has highest forecasting precision than the other three models, and its average prediction error is about 7.26%.The distance measuring mode with exponent weight has higher accuracy in choosing the nearest neighbors, and its average prediction error is about 5.57%.The predicting performance of improved K-NN model with exponent weight and space-time parameters is best compared with the artificial neural network model and the historical average model, and its average prediction error is only 9.43%.So the improved K-NN model is an effective way for forecasting short-term traffic flow.
- traffic information engineering /
- short-term traffic flow forecast /
- K-nearest neighbor model /
- space-time parameters /
- exponent weight

HTML全文

图 1 基于时间维度的预测机理

Figure 1. Prediction mechanism based on time dimension

下载: 全尺寸图片幻灯片

图 2 基于空间维度的预测机理

Figure 2. Prediction mechanism based on space dimension

下载: 全尺寸图片幻灯片

图 3 相关系数权重法

Figure 3. Correlation coefficient weighting method

下载: 全尺寸图片幻灯片

图 4 指数权重法

Figure 4. Exponent weighting method

下载: 全尺寸图片幻灯片

图 5 研究路段信息

Figure 5. Informations of research road sections

下载: 全尺寸图片幻灯片

图 6 GPS系统定位的出租车点

Figure 6. GPS positioned taxis

下载: 全尺寸图片幻灯片

图 7 三个路段不同时段的平均速度

Figure 7. Average speeds of three road sections during different times

下载: 全尺寸图片幻灯片

图 8 状态向量对相对预测误差的影响

Figure 8. Influences of state vectors on relative prediction errors

下载: 全尺寸图片幻灯片

图 9 距离度量方式对相对预测误差的影响

Figure 9. Influences of distance measure modes on relative prediction errors

下载: 全尺寸图片幻灯片

图 10 不同模型的预测结果比较

Figure 10. Comparison of prediction results of different models

下载: 全尺寸图片幻灯片

表 1 四种状态向量

Table 1. Four state vectors

下载: 导出CSV

表 2 某出租车在中华路段1上的GPS数据

Table 2. GPS data of a certain taxi on Zhonghua Road 1

下载: 导出CSV

参考文献(16)

[1]	张晓利, 陆化普. 非参数回归方法在短时交通流预测中的应用[J]. 清华大学学报: 自然科学版, 2009, 49(9): 1471-1475. doi: 10.3321/j.issn:1000-0054.2009.09.011 ZHANG Xiao-li, LU Hua-pu. Non-parametric regression and application for short-term traffic flow forecasting[J]. Journal of Tsinghua University: Science and Technology, 2009, 49(9): 1471-1475. (in Chinese). doi: 10.3321/j.issn:1000-0054.2009.09.011
[2]	刘静, 关伟. 交通流预测方法综述[J]. 公路交通科技, 2004, 21(3): 82-85. https://www.cnki.com.cn/Article/CJFDTOTAL-GLJK2020S1025.htm LIU Jing, GUAN Wei. A summary of traffic flow forecasting methods[J]. Journal of Highway and Transportation Research and Development, 2004, 21(3): 82-85. (in Chinese). https://www.cnki.com.cn/Article/CJFDTOTAL-GLJK2020S1025.htm
[3]	朱顺应, 王红, 李关寿. 路段上短时间区段内交通量预测ARIMA模型[J]. 重庆交通学院学报, 2003, 22(1): 76-77, 95. https://www.cnki.com.cn/Article/CJFDTOTAL-CQJT200301019.htm ZHU Shun-ying, WANG Hong, LI Guan-shou. The ARIMA model used in forecasting of traffic volume in short interval on the link[J]. Journal of Chongqing Jiaotong University, 2003, 22(1): 76-77, 95. (in Chinese). https://www.cnki.com.cn/Article/CJFDTOTAL-CQJT200301019.htm
[4]	OKUTANI I, STEPHANEDES Y J. Dynamic prediction of traffic volume through Kalman filtering theory[J]. Transportation Research Part B: Methodological, 1984, 18(l): 1-11.
[5]	聂佩林, 余志, 何兆成. 基于约束卡尔曼滤波的短时交通流量组合预测模型[J]. 交通运输工程学报, 2008, 8(5): 86-90. http://transport.chd.edu.cn/article/id/200805017 NIE Pei-lin, YU Zhi, HE Zhao-cheng. Constrained Kalman filter combined predictor for short-term traffic flow[J]. Journal of Traffic and Transportation Engineering, 2008, 8(5): 86-90. (in Chinese). http://transport.chd.edu.cn/article/id/200805017
[6]	宋国杰, 胡程, 谢昆青, 等. 面向实时短时交通流预测的过程神经元网络建模[J]. 交通运输工程学报, 2009, 9(5): 73-77. http://transport.chd.edu.cn/article/id/200905013 SONG Guo-jie, HU Cheng, XIE Kun-qing, et al. Process neural network modeling for real time short-term traffic flow prediction[J]. Journal of Traffic and Transportation Engineering, 2009, 9(5): 73-77. (in Chinese). http://transport.chd.edu.cn/article/id/200905013
[7]	SMITH B L, WILLIAMS B M, OSWALD R K. Comparison of parametric and nonparametric models for traffic flow forecasting[J]. Transportation Research Part C: Emerging Technologies, 2002, 10(4): 303-321. doi: 10.1016/S0968-090X(02)00009-8
[8]	HSU C W, LIN C J. A comparison of methods for multi-class support vector machines[J]. IEEE Transactions on Neural Networks, 2002, 13(2): 415-425. doi: 10.1109/72.991427
[9]	徐启华, 杨瑞. 支持向量机在交通流量实时预测中的应用[J]. 公路交通科技, 2005, 22(12): 131-134. https://www.cnki.com.cn/Article/CJFDTOTAL-GLJK200512033.htm XU Qi-hua, YANG Rui. Traffic flow prediction using sup-port vector machine based method[J]. Journal of Highway and Transportation Research and Development, 2005, 22(12): 131-134. (in Chinese). https://www.cnki.com.cn/Article/CJFDTOTAL-GLJK200512033.htm
[10]	干宏程, 汪晴, 范炳全. 基于宏观交通流模型的行程时间预测[J]. 上海理工大学学报, 2008, 30(5): 409-413. https://www.cnki.com.cn/Article/CJFDTOTAL-HDGY200805002.htm GAN Hong-cheng, WANG Qing, FAN Bing-quan. Travel time prediction based on macroscopic traffic flow models[J]. Journal of University of Shanghai for Science and Technology, 2008, 30(5): 409-413. (in Chinese). https://www.cnki.com.cn/Article/CJFDTOTAL-HDGY200805002.htm
[11]	翁剑成, 荣健, 任福田, 等. 基于非参数回归的快速路行程速度短期预测算法[J]. 公路交通科技, 2007, 24(3): 93-97, 106. https://www.cnki.com.cn/Article/CJFDTOTAL-GLJK200703021.htm WENG Jian-cheng, RONG Jian, REN Fu-tian, et al. Non-parametric regression model based short-term prediction for expressway travel speed[J]. Journal of Highway and Trans-portation Research and Development, 2007, 24(3): 93-97, 106. (in Chinese). https://www.cnki.com.cn/Article/CJFDTOTAL-GLJK200703021.htm
[12]	张晓利, 贺国光, 陆化普. 基于K-邻域非参数回归短时交通流预测方法[J]. 系统工程学报, 2009, 24(2): 178-183. https://www.cnki.com.cn/Article/CJFDTOTAL-XTGC200902009.htm ZHANG Xiao-li, HE Guo-guang, LU Hua-pu. Short-term traffic flow forecasting based on K-nearest neighbors non-parametric regression[J]. Journal of Systems Engineering, 2009, 24(2): 178-183. (in Chinese). https://www.cnki.com.cn/Article/CJFDTOTAL-XTGC200902009.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.
[14]	张涛, 陈先, 谢美萍, 等. 基于K近邻非参数回归的短时交通流预测方法[J]. 系统工程理论与实践, 2010, 30(2): 376-384. https://www.cnki.com.cn/Article/CJFDTOTAL-XTLL201002026.htm ZHANG Tao, CHEN Xian, XIE Mei-ping, et al. K-NN based nonparametric regression method for short-term traffic flow forecasting[J]. Systems Engineering—Theory and Practice, 2010, 30(2): 376-384. (in Chinese). https://www.cnki.com.cn/Article/CJFDTOTAL-XTLL201002026.htm
[15]	SMITH B L, DEMETSKY M J. Traffic flow forecasting: comparison of modeling approaches[J]. Journal of Transpor-tation Engineering, 1997, 123(4): 261-266.
[16]	YU Bin, LAM W H K, TAM M L. Bus arrival time predic-tion at bus stop with multiple routes[J]. Transportation Research Part C: Emerging Technologies, 2011, 19(6): 1157-1170.