吉林大学学报(工学版) ›› 2021, Vol. 51 ›› Issue (4): 1241-1250.doi: 10.13229/j.cnki.jdxbgxb20200288

• 交通运输工程·土木工程 • 上一篇    

基于KNN回归的客运枢纽聚集人数组合预测方法

卢凯1,2(),吴蔚1,林观荣3(),田鑫1,徐建闽1,2   

  1. 1.华南理工大学 土木与交通学院,广州 510640
    2.现代城市交通技术江苏高校协同创新中心,南京 210096
    3.深圳平安信息技术有限公司,广东 深圳 518052
  • 收稿日期:2020-04-30 出版日期:2021-07-01 发布日期:2021-07-14
  • 通讯作者: 林观荣 E-mail:kailu@scut.edu.cn;linguanrong@sutpc.com
  • 作者简介:卢凯(1979-),男,教授,博士. 研究方向:交通信息工程及控制.E-mail:kailu@scut.edu.cn
  • 基金资助:
    国家自然科学基金项目(61773168);中央高校基本科研业务费专项项目(2019ZD45)

Combination forecasting model for number of assembling passengers at transportation terminal based on KNN regression algorithm

Kai LU1,2(),Wei WU1,Guan-rong LIN3(),Xin TIAN1,Jian-min XU1,2   

  1. 1.School of Civil Engineering and Transportation,South China University of Technology,Guangzhou 510640,China
    2.Jiangsu Province Collaborative Innovation Center of Modern Urban Traffic Technologies,Nanjing 210096,China
    3.Shenzhen Ping An Information Technology Co. ,Ltd. ,Shenzhen 518052,China
  • Received:2020-04-30 Online:2021-07-01 Published:2021-07-14
  • Contact: Guan-rong LIN E-mail:kailu@scut.edu.cn;linguanrong@sutpc.com

摘要:

为精准预测客运枢纽聚集人数以形成合理科学的客运枢纽客流组织方案,提出了一种基于KNN回归算法的客运枢纽聚集人数组合预测方法。在分析客运枢纽客流聚集规律的基础上,以数值相似和趋势相似为原则运用KNN回归算法预测区域聚集人数,并综合考虑各自特点引入时变权重系数进行组合预测,解决了以往KNN回归预测模型所需历史数据量大和运行时间长等方面的不足。实例分析结果表明,本文方法在非节假日平均预测精度可达95%以上,在春运期间平均预测精度可达90%,均高于移动平均法、卡尔曼滤波法与灰色预测法。

关键词: 交通运输系统工程, 客运枢纽, KNN回归, 组合预测, 状态向量, 时变权重

Abstract:

To work out a reasonable and scientific passenger-flow organization scheme with an accurate assembling passenger prediction for transportation terminals, a combination forecasting model based on the KNN regression algorithm was proposed. Grounded on the analysis of the assembling laws of passengers at transportation terminals, the KNN regression algorithm was applied to forecast the number of assembling passengers based on the principles of the numerical similarity and the trend similarity. With the comprehensive consideration of the respective characteristics, the combination forecasting was realized by introducing a time-varying weight coefficient. As a result, the proposed model could solve the shortcomings of the previous KNN regression prediction model, such as large amount of historical data and long running time. The experimental results suggest that the average prediction accuracy of the proposed method can be guaranteed over 95% during non-holidays and 90% during the Spring Festival travel rush, which is superior to moving average method, Kalman filter model and gray prediction method respectively.

Key words: engineering of transportation system, transportation terminal, k-nearest neighbor regression, combination forecasting, state vector, time-varying weight

中图分类号: 

  • U491

图1

广州火车站聚集人数变化"

图2

KNN预测原理示意图"

图3

算法流程图"

图4

时变权重系数"

图5

参数的选取"

图6

不同预测时长的平均预测误差"

图7

不同时变权重系数的平均预测误差"

图8

3种KNN回归方法预测结果性能指标MAPE"

图9

非节假日聚集人数预测结果"

图10

春运期间聚集人数预测结果"

表1

四种方法在非节假日的预测结果性能指标MAPE (%)"

日期11/711/811/911/1011/1111/1211/13均值
灰色预测法7.805.333.247.155.534.574.485.44
卡尔曼滤波法11.245.993.127.086.324.285.026.15
移动平均法8.153.253.406.836.306.724.675.62
KNN组合6.455.252.746.325.192.944.884.83

表2

四种方法在春运期间的预测结果性能指标MAPE (%)"

日期灰色预测法卡尔曼滤波法移动平均法KNN组合法日期灰色预测法卡尔曼滤波法移动平均法KNN组合法
2/123.6822.8433.557.082/2140.4331.0126.7919.52
2/215.5113.9429.514.772/2232.8923.678.9515.83
2/311.019.7725.217.532/2320.3520.6812.6113.47
2/47.105.9321.383.972/2419.6012.347.5016.99
2/57.235.5019.343.952/2516.228.318.8013.38
2/67.494.5415.814.522/2612.166.558.498.37
2/77.644.3113.923.652/278.848.6611.286.07
2/810.094.859.617.212/2812.6612.1412.384.99
2/911.535.247.815.863/115.2411.7713.478.04
2/1011.774.707.994.603/29.164.606.069.73
2/1111.922.617.372.553/319.328.5213.506.95
2/1225.2111.377.017.893/417.275.9614.946.14
2/1346.4528.2120.5813.733/522.325.4718.474.35
2/1489.5362.5254.387.573/620.495.798.048.17
2/15313.18242.01117.9833.203/719.947.188.626.06
2/16255.59174.608.6717.353/816.6712.146.0516.05
2/17150.6174.5564.096.023/924.076.236.625.42
2/18121.1935.8527.568.023/1024.416.337.327.38
2/1992.7910.2329.2610.023/1128.865.2114.416.16
2/2069.7515.7931.2814.243/1230.905.4217.498.73

图11

预测方法的性能指标(MAPE)"

1 耿美君. 综合客运枢纽服务水平评价研究[D]. 长春: 吉林大学交通学院, 2009.
Geng Mei-jun. Research on LOS evaluation of comprehensive passenger transport terminal[D]. Changchun: College of Transportation,Jilin University, 2009.
2 Cover T M, Hart P E. Nearest neighbor pattern classification[J]. IEEE Transactions on Information Theory, 1967, 13(1):21-27.
3 Yakowitz S. Nearest-neighbour methods for time series analysis[J]. Journal of Time Series Analysis,1987: 8(2): 235-247.
4 Davis G A, Nihan N. Nonparametric regression and short-term freeway traffic forecasting[J]. Journal of Transportation Engineering, 1991, 117(2): 178-188.
5 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.
Zhang Tao, Chen Xian, Xie Mei-ping, et al. KNN based nonparametric regression method for short-term traffic flow forecasting[J]. Systems Engineering-Theory & Practice, 2010, 30(2):376-384.
6 豆飞, 贾利民, 秦勇, 等. 铁路客运专线模糊k近邻客流预测模型[J]. 中南大学学报:自然科学版, 2014, 45(12):4422-4430.
Dou Fei, Jia Li-min, Qin Yong, et al. Fuzzy k-nearest neighbor passenger flow forecasting model of passenger dedicated line[J]. Journal of Central South University (Science and Technology), 2014, 45(12):4422-4430.
7 林培群, 陈丽甜, 雷永巍. 基于K近邻模式匹配的地铁客流量短时预测[J]. 华南理工大学学报:自然科学版, 2018, 46(1):50-57.
Lin Pei-qun, Chen Li-tian, Lei Yong-wei. Short-term forecasting of subway traffic based on K-nearest neighbour pattern matching[J]. Journal of South China University of Technology (Natural Science Edition), 2018, 46(1):50-57.
8 姚恩建, 周文华, 张永生. 城市轨道交通新站开通初期实时进出站客流量预测[J]. 中国铁道科学, 2018, 39(2):119-127.
Yao En-jian, Zhou Wen-hua, Zhang Yong-sheng. Real-time forecast of entrance and exit passenger flow for newly opened station of urban rail transit at initial stage[J]. China Railway Science, 2018, 39(2):119-127.
9 郇宁, 谢俏, 叶红霞, 等. 基于改进KNN算法的城轨进站客流实时预测[J]. 交通运输系统工程与信息, 2018, 18(5):121-128.
Huan Ning, Xie Qiao, Ye Hong-xia, et al. Real-time forecasting of urban rail transit ridership at the station level based on improved KNN algorithm[J]. Journal of Transportation Systems Engineering and Information Technology, 2018, 18(5):121-128.
10 邢志伟, 何川, 罗谦, 等. 基于双层K近邻算法航站楼短时客流量预测[J]. 北京航空航天大学学报, 2019, 45(1): 26-34
Xing Zhi-wei, He Chuan, Luo Qian, et al. Terminal building short-term passenger flow forecast based on two-tier K-nearest neighbor algorithm[J]. Journal of Beijing University of Aeronautics and Astronautics, 2019,45(1):26-34.
11 赵亚伟, 陈艳晶. 多维时间序列的组合预测模型[J]. 中国科学院大学学报, 2016, 33(6):825-833.
Zhao Ya-wei, Chen Yan-jing. A combined prediction model for multi-dimensional time series[J]. Journal of University of Chinese Academy of Sciences, 2016, 33(6):825-833.
12 刘欣彤, 黄小龙, 谢秉磊. 基于SVM-KNN的降雨条件下短时公交客流预测[J]. 交通信息与安全, 2018, 36(5):117-123.
Liu Xin-tong, Huang Xiao-long, Xie Bing-lei. A model of short-term forecast of passenger flow of buses based on SVM-KNN under rainy conditions[J]. Journal of Transport Information and Safety, 2018, 36(5):117-123.
13 孟品超, 李学源, 贾洪飞, 等. 基于滑动平均法的轨道交通短时客流实时预测[J]. 吉林大学学报:工学版, 2018,48(2):448-453.
Meng Pin-chao, Li Xue-yuan, Jia Hong-fei, et al. Short-time rail transit passenger flow real-time prediction based on moving average[J]. Journal of Jilin University (Engineering and Technology Edition), 2018, 48(2): 448-453.
[1] 彭博,张媛媛,王玉婷,唐聚,谢济铭. 基于自动编码机-分类器的视频交通状态自动识别[J]. 吉林大学学报(工学版), 2021, 51(3): 886-892.
[2] 张健,吴坤润,杨敏,冉斌. 智能网联环境下交叉口双环自适应控制模型[J]. 吉林大学学报(工学版), 2021, 51(2): 541-548.
[3] 王殿海,沈辛夷,罗小芹,金盛. 车均延误最小情况下的相位差优化方法[J]. 吉林大学学报(工学版), 2021, 51(2): 511-523.
[4] 宋现敏,张明业,李振建,王鑫,张亚南. 动态公交专用道的设置及其仿真分析评价[J]. 吉林大学学报(工学版), 2020, 50(5): 1677-1686.
[5] 贾洪飞,丁心茹,杨丽丽. 城市潮汐车道优化设计的双层规划模型[J]. 吉林大学学报(工学版), 2020, 50(2): 535-542.
[6] 尹超英,邵春福,王晓全,熊志华. 考虑空间异质性的建成环境对通勤方式选择的影响[J]. 吉林大学学报(工学版), 2020, 50(2): 543-548.
[7] 张大伟,祝海涛. 考虑行人差异性的人群疏散最优决策理论模型[J]. 吉林大学学报(工学版), 2020, 50(2): 549-556.
[8] 常玉林,袁才鸿,孙超,张鹏. 基于改进元胞传输模型的城市路网实际阻抗计算方法[J]. 吉林大学学报(工学版), 2020, 50(1): 132-139.
[9] 宗长富,文龙,何磊. 基于欧几里得聚类算法的三维激光雷达障碍物检测技术[J]. 吉林大学学报(工学版), 2020, 50(1): 107-113.
[10] 隋振,姜源. 基于MIMO类脑情感学习回路的横-纵向综合控制驾驶员模型[J]. 吉林大学学报(工学版), 2020, 50(1): 140-146.
[11] 别一鸣,姜凯,汤茹茹,王琳虹,熊昕宇. 考虑方案过渡影响的单点交通控制时段划分方法[J]. 吉林大学学报(工学版), 2019, 49(6): 1844-1851.
[12] 谷远利, 张源, 芮小平, 陆文琦, 李萌, 王硕. 基于免疫算法优化LSSVM的短时交通流预测[J]. 吉林大学学报(工学版), 2019, 49(6): 1852-1857.
[13] 程国柱, 冯思鹤, 冯天军. 占用车行道的路内停车泊位设置条件[J]. 吉林大学学报(工学版), 2019, 49(6): 1858-1864.
[14] 龙海波,杨家其,赵学彧. 基于转运延误风险的多方式协同货运载运工具配置优化[J]. 吉林大学学报(工学版), 2019, 49(5): 1492-1499.
[15] 别一鸣,汤茹茹,王运豪,文斌,冯天军,王琳虹. 信号交叉口进口车道饱和流率估计方法[J]. 吉林大学学报(工学版), 2019, 49(5): 1459-1464.
Viewed
Full text


Abstract

Cited

  Shared   
  Discussed   
No Suggested Reading articles found!