基于卷积长短时记忆网络的短时公交客流量预测

doi:10.16182/j.issn1004731x.joss.22-1168

系统仿真学报 ›› 2024, Vol. 36 ›› Issue (2): 476-486.doi: 10.16182/j.issn1004731x.joss.22-1168

基于卷积长短时记忆网络的短时公交客流量预测

陈静¹(), 张昭冲¹(), 王琳凯¹, 安脉², 王伟¹

^1.天津职业技术师范大学信息技术工程学院，天津 300222
^2.中新天津生态城管委会智慧城市发展局，天津 300467

收稿日期:2022-10-04 修回日期:2022-12-08 出版日期:2024-02-15 发布日期:2024-02-04
通讯作者: 张昭冲 E-mail:c_j_223@163.com;s789658@qq.com
第一作者简介:陈静(1984-)，女，副教授，硕士生导师，博士，研究方向为机器学习。E-mail：c_j_223@163.com
基金资助:
天津市教委科研计划(2021KJ008)

Short-term Bus Passenger Flow Prediction Based on Convolutional Long-short-term Memory Network

Chen Jing¹(), Zhang Zhaochong¹(), Wang Linkai¹, An Mai², Wang Wei¹

^1.School of Information Technology and Engineering, Tianjin University of Technology and Education, Tianjin 300222, China
^2.Smart City Development Authority, Sino-Singapore Tianjin Eco-city Management Committee, Tianjin 300467, China

Received:2022-10-04 Revised:2022-12-08 Online:2024-02-15 Published:2024-02-04
Contact: Zhang Zhaochong E-mail:c_j_223@163.com;s789658@qq.com

摘要/Abstract

摘要：

针对传统的短时客流预测方法没有考虑到时序特征中跨时段客流之间的相似性问题，提出一种改进k-means聚类算法与卷积神经网络和长短时记忆网络相结合的短时客流量预测模型k-CNN-LSTM。通过k-means算法对跨时段时序数据进行聚类，使用间隔统计确定k值，构建交通流矩阵模型，采用CNN-LSTM网络处理具有时空特征的短时客流。该模型能够对具有空间相关性的数据进行较为准确的预测。使用真实数据集对模型进行检验和参数调优，实验结果表明：k-CNN-LSTM模型较其他模型有相对较高的预测精度。

关键词: 卷积神经网络, 长短时记忆网络, 时空数据预测, k-means聚类, 客流量预测

Abstract:

To address the problem that the traditional short-time passenger flow prediction method does not consider the temporal characteristics similarity between the inter-temporal passenger flows, a short-time passenger flow prediction model k-CNN-LSTM is proposed by combining the improved k-means clustering algorithm with the CNN and the LSTM. The k-means is used to cluster the intertemporal time-series data, the k-value is determined by using the gap-statistic, and a traffic flow matrix model is constructed. A CNN-LSTM network is used to process the short-time passenger flows with spatial and temporal characteristics. The model is tested and parameter tuned by the real dataset. The test results show the model is able to predict the spatially correlated data more accurately.

Key words: CNN, LSTM, spatiotemporal data prediction, k-means clustering, passenger flow prediction

中图分类号:

TP181

陈静,张昭冲,王琳凯等 . 基于卷积长短时记忆网络的短时公交客流量预测[J]. 系统仿真学报, 2024, 36(2): 476-486.

Chen Jing,Zhang Zhaochong,Wang Linkai,et al . Short-term Bus Passenger Flow Prediction Based on Convolutional Long-short-term Memory Network[J]. Journal of System Simulation, 2024, 36(2): 476-486.

图/表 15

图1

图2

图3

图4

图5

图6

图7

图8

图9

表1

图10

图11

图12

图13

表2

参考文献 21

1	谢振东, 刘雪琴, 吴金成, 等. 公交IC卡数据客流预测模型研究[J]. 广东工业大学学报, 2018, 35(1): 16-22.
	Xie Zhendong, Liu Xueqin, Wu Jincheng, et al. A Study of Passenger Flow Prediction Based on IC Card Data[J]. Journal of Guangdong University of Technology, 2018, 35(1): 16-22.
2	赵鹏, 李璐. 基于ARIMA模型的城市轨道交通进站量预测研究[J]. 重庆交通大学学报(自然科学版), 2020, 39(1): 40-44.
	Zhao Peng, Li Lu. Prediction of Urban Rail Transit Station Inflows Based on ARIMA Model[J]. Journal of Chongqing Jiaotong University(Natural Science), 2020, 39(1): 40-44.
3	张春辉, 宋瑞, 孙杨. 基于卡尔曼滤波的公交站点短时客流预测[J]. 交通运输系统工程与信息, 2011, 11(4): 154-159.
	Zhang Chunhui, Song Rui, Sun Yang. Kalman Filter-based Short-term Passenger Flow Forecasting on Bus Stop[J]. Journal of Transportation Systems Engineering and Information Technology, 2011, 11(4): 154-159.
4	Yang Zhenqi. Highway Traffic Accident Prediction Based on SVR Trained by Genetic Algorithm[J]. Advanced Materials Research, 2012, 433-440: 5886-5889.
5	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 Xintong, Huang Xiaolong, Xie Binglei. 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.
6	Chen Quanchao, Wen Di, Li Xuqiang, et al. Empirical Mode Decomposition Based Long Short-term Memory Neural Network Forecasting Model for the Short-term Metro Passenger Flow[J]. PloS one, 2019, 14(9): e0222365.
7	戢晓峰, 戈艺澄. 基于深度学习的节假日高速公路交通流预测方法[J]. 系统仿真学报, 2020, 32(6): 1164-1171.
	Ji Xiaofeng, Ge Yicheng. Holiday Highway Traffic Flow Prediction Method Based on Deep Learning[J]. Journal of System Simulation, 2020, 32(6): 1164-1171.
8	刘明宇, 吴建平, 王钰博, 等. 基于深度学习的交通流量预测[J]. 系统仿真学报, 2018, 30(11): 4100-4105, 4114.
	Liu Mingyu, Wu Jianping, Wang Yubo, et al. Traffic Flow Prediction Based on Deep Learning[J]. Journal of System Simulation, 2018, 30(11): 4100-4105, 4114.
9	Zhao Zheng, Chen Weihai, Wu Xingming, et al. LSTM Network: A Deep Learning Approach for Short-term Traffic Forecast[J]. IET Intelligent Transport Systems, 2017, 11(2): 68-75.
10	Cao Miaomiao, Li V O K, Chan V W S. A CNN-LSTM Model for Traffic Speed Prediction[C]//2020 IEEE 91st Vehicular Technology Conference (VTC2020-Spring). Piscataway, NJ, USA: IEEE, 2020: 1-5.
11	Bogaerts Toon, Masegosa Antonio D, Angarita-Zapata Juan S, et al. A Graph CNN-LSTM Neural Network for Short and Long-term Traffic Forecasting Based on Trajectory Data[J]. Transportation Research Part C: Emerging Technologies, 2020, 112: 62-77.
12	龚兰兰, 凌兴宏. CNN+GRU公交短时客流预测研究[J]. 现代电子技术, 2021, 44(14): 70-74.
	Gong Lanlan, Ling Xinghong. Research on Short-term Bus Passenger Flow Prediction Based on CNN+GRU[J]. Modern Electronics Technique, 2021, 44(14): 70-74.
13	王祥雪, 许伦辉. 基于深度学习的短时交通流预测研究[J]. 交通运输系统工程与信息, 2018, 18(1): 81-88.
	Wang Xiangxue, Xu Lunhui. Short-term Traffic Flow Prediction Based on Deep Learning[J]. Journal of Transportation Systems Engineering and Information Technology, 2018, 18(1): 81-88.
14	谢英豪, 江志彬, 徐瑞华. 城市轨道交通断面客流数据可视化分析及优化[J]. 城市轨道交通研究, 2017, 20(1): 69-74.
	Xie Yinghao, Jiang Zhibin, Xu Ruihua. Optimized Visualization Analysis of the Section Passenger Data in Rail Transit Network[J]. Urban Mass Transit, 2017, 20(1): 69-74.
15	Li Runmei, Huang Yinfeng, Wang Jian. Long-term Traffic Volume Prediction Based on K-means Gaussian Interval Type-2 Fuzzy sets[J]. IEEE/CAA Journal of Automatica Sinica, 2019, 6(6): 1344-1351.
16	陈振祥, 林培杰, 程树英, 等. 基于K-means++和混合深度学习的光伏功率预测[J]. 电气技术, 2021, 22(9): 7-13, 33.
	Chen Zhenxiang, Lin Peijie, Cheng Shuying, et al. Photovoltaic Power Prediction Based on K-means++ and Hybrid Deep Learning[J]. Electrical Engineering, 2021, 22(9): 7-13, 33.
17	Tibshirani R, Walther G, Hastie T. Estimating the Number of Clusters in a Data Set Via the Gap Statistic[J]. Journal of the Royal Statistical Society Series B: Statistical Methodology, 2001, 63(2): 411-423.
18	陈东洋, 陈德旺, 江世雄, 等. 基于时段客流特征聚类的地铁运营时段划分[J]. 计算机系统应用, 2021, 30(3): 256-261.
	Chen Dongyang, Chen Dewang, Jiang Shixiong, et al. Division of Metro Operation Periods Based on Feature Clustering of Passenger Flow[J]. Computer Systems & Applications, 2021, 30(3): 256-261.
19	Chen Ken, Chen Fei, Lai Baisheng, et al. Dynamic Spatio-temporal Graph-based CNNs for Traffic Flow Prediction[J]. IEEE Access, 2020, 8: 185136-185145.
20	孙越. 组合模型在铁路客流量预测中的应用研究[D]. 兰州: 兰州交通大学, 2020.
	Sun Yue. The Applications and Research of Combined Model in Railway Passenger Flow Forecast[D]. Lanzhou: Lanzhou Jiaotong University, 2020.
21	杨丽, 吴雨茜, 王俊丽, 等. 循环神经网络研究综述[J]. 计算机应用, 2018, 38(增2): 1-6, 26.
	Yang Li, Wu Yuxi, Wang Junli, et al. Research on Recurrent Neural Network[J]. Journal of Computer Applications, 2018, 38(S2): 1-6, 26.

模型名称	参数名称	参数值
k-CNN-LSTM	B/E/OPT/LR/A/LOSS	128/300/Adam/0.01/ReLU/MSE
CNN-LSTM	B/E/OPT/LR/A/LOSS	128/300/Adam/0.01/ReLU/MSE
BiLSTM	B/E/OPT/LR/LOSS	128/300/Adam/0.01/MSE
LSTM	B/E/OPT/LR/LOSS	128/300/Adam/0.01/MSE
SVR	Kernel/CV/LOSS	Rbf/10/MSE

模型	商业街			服务中心			第三社区中心
模型	MSE	MAE	SMAPE/%	MSE	MAE	SMAPE/%	MSE	MAE	SMAPE/%
k-CNN-LSTM	227.48	11.30	45.44	164.60	9.77	45.16	150.44	9.37	47.26
CNN-LSTM	280.55	12.15	48.39	217.99	10.79	49.94	186.23	10.34	50.87
Bi-LSTM	271.56	12.10	48.13	191.76	11.18	54.49	168.11	10.19	48.00
LSTM	311.93	12.67	47.85	198.88	10.85	50.98	283.36	13.68	55.68
SVR	252.74	12.90	50.77	164.93	9.82	45.80	160.00	10.27	49.57

[1]	韩莹, 王乐豪, 王淑梅, 张翔, 罗星星. 宽度-深度融合时频分析的径流智能预测方法[J]. 系统仿真学报, 2024, 36(2): 363-372.
[2]	徐艺博, 于清华, 王炎娟, 郭策, 冯世如, 卢惠民. 基于多源信息融合的巡飞弹对地目标识别与毁伤评估[J]. 系统仿真学报, 2024, 36(2): 511-521.
[3]	张凯, 卢海鹏, 韩莹, 张龄允, 丁昱杰. 融合变分模态分解的时空卷积短时车速预测[J]. 系统仿真学报, 2023, 35(8): 1651-1660.
[4]	杨静, 陆铭华, 胡星辰, 吴金平. 基于不确定时间序列的来袭目标解算方法[J]. 系统仿真学报, 2023, 35(6): 1245-1259.
[5]	王旭, 季伟东, 周国辉, 杨佳慧. 基于多指标精英个体博弈机制的多目标优化算法[J]. 系统仿真学报, 2023, 35(3): 494-514.
[6]	蔡兴泉, 李治均, 奚梦瑶, 孙海燕. 基于神经网络的手绘服饰图纹上色及风格迁移[J]. 系统仿真学报, 2023, 35(3): 604-615.
[7]	杨丰玮, 陈鹏, 郗凯, 蒲华林, 刘雪垠. 融合空间信息的运动想象脑电在线分类方法[J]. 系统仿真学报, 2023, 35(2): 254-267.
[8]	张立峰, 王会忍. 基于卷积神经网络及有限元仿真的电容层析成像图像重建[J]. 系统仿真学报, 2022, 34(4): 712-718.
[9]	敖邦乾, 杨莎, 令狐金卿, 叶振环. 基于级联神经网络疲劳驾驶检测系统设计[J]. 系统仿真学报, 2022, 34(2): 323-333.
[10]	王亚茹, 杨凯, 翟永杰, 郭聪彬, 赵文清, 苏杰. 基于人工图像数据扩充的输电线路绝缘子识别[J]. 系统仿真学报, 2022, 34(11): 2337-2347.
[11]	徐佳乐, 张海东, 赵东海, 倪晚成. 基于卷积神经网络的陆战兵棋战术机动策略学习[J]. 系统仿真学报, 2022, 34(10): 2181-2193.
[12]	郭业才, 王庆伟. 基于截断迁移与并行残差网络的调制识别算法[J]. 系统仿真学报, 2022, 34(09): 2009-2018.
[13]	林硕, 安磊, 高治军, 单丹, 尚文利. 结合栈式自编码及长短时记忆的入侵检测研究[J]. 系统仿真学报, 2021, 33(6): 1288-1296.
[14]	赵瑛, 陆耀, 张健, 梁启弟, 龙炜. 基于深度神经网络的多视角人体动作识别[J]. 系统仿真学报, 2021, 33(5): 1019-1030.
[15]	仝卫国, 庞雪纯, 朱赓宏. 基于卷积神经网络的气液两相流流型识别方法[J]. 系统仿真学报, 2021, 33(4): 883-891.

基于卷积长短时记忆网络的短时公交客流量预测

Short-term Bus Passenger Flow Prediction Based on Convolutional Long-short-term Memory Network

RichHTML

PDF (PC)

可视化

摘要/Abstract

引用本文

使用本文

图/表 15

参考文献 21

相关文章 15

编辑推荐

Metrics

本文评价