基于双向长短期记忆网络的公交到站时间预测模型

doi:10.16097/j.cnki.1009-6744.2023.02.016

交通运输系统工程与信息 ›› 2023, Vol. 23 ›› Issue (2): 148-160.DOI: 10.16097/j.cnki.1009-6744.2023.02.016

基于双向长短期记忆网络的公交到站时间预测模型

张兵^*1，周丹丹¹，孙健²，倪训友¹

1. 华东交通大学，交通运输工程学院，南昌 330013；2. 长安大学，未来交通学院，西安 710064

收稿日期:2022-12-11 修回日期:2023-01-26 接受日期:2023-02-07 出版日期:2023-04-25 发布日期:2023-04-19
作者简介:张兵(1981- )，男，山东济宁人，副教授，博士
基金资助:
国家自然科学基金(71971138, 52162042, 72161012)

Bus Arrival Time Prediction Model Based on Bidirectional Long Short-term Memory Network

ZHANG Bing^*1, ZHOU Dan-dan¹, SUN Jian², NI Xun-you¹

1. School of Transportation Engineering, East China Jiaotong University, Nanchang 330013, China; 2. College of Future Transportation, Chang'an University, Xi'an 710064, China

Received:2022-12-11 Revised:2023-01-26 Accepted:2023-02-07 Online:2023-04-25 Published:2023-04-19
Supported by:
National Natural Science Foundation of China (71971138, 52162042, 72161012)

摘要/Abstract

摘要： 为实现准确预测公交到站时间，提高城市公交出行分担率，本文提出一种基于双向长短期记忆神经网络(BiLSTM)并考虑超参数寻优的公交到站时间预测模型。通过引入非线性收敛因子、正弦余弦算子及自适应参数改进海鸥算法对双向 LSTM 模型实现超参数寻优，并增加Attention机制以提高双向LSTM处理信息能力，构建改进海鸥算法优化增加Attention机制的双向LSTM(ISOA-BiLSTM-Attention)预测模型。使用中国江西省南昌市220路公交GPS数据分方向和分时段预测车辆到站时间，验证模型预测精度。结果表明：整体上来说，Attention机制优化后的双向LSTM模型比单独采用双向LSTM模型预测精度更好；进的海鸥算法可对双向LSTMAttention 模型实现较好的优化效果，相较于现有模型及标准海鸥算法(SOA)优化双向LSTMAttention模型，ISOA-BiLSTM-Attention对于不同方向及不同时段公交到站时间预测的平均绝对百分比误差、均方根误差及平均绝对误差至少分别降低 5.96%、9.87%及 7.99%；同时，ISOABiLSTM-Attention具有最大的模型决定系数 R2 值体现了该预测模型泛化能力及稳定性较好，可针对公交到站时间进行较为准确地拟合。

关键词: 城市交通, 公交到站时间预测, 改进海鸥优化算法, 双向LSTM模型, Attention机制

Abstract: To improve the accuracy of bus arrival time prediction and increase the bus usage in the cities, this paper proposes a bus arrival time prediction model based on a bidirectional Long Short-term Memory (BiLSTM) neural network and the hyperparameter search. The improved seagull algorithm optimization adding Attention mechanism to bidirectional LSTM (ISOA-BiLSTM-Attention) prediction model was developed by introducing nonlinear convergence factor, sine cosine operator, and adaptive parameters to improve the seagull algorithm to achieve hyperparametric optimization of the bidirectional LSTM model. The Attention mechanism was added to improve the information processing ability of bidirectional LSTM. Then, the trajectory data of bus route 220 in Nanchang, Jiangxi Province of China, were used to predict the bus arrival time for different directions and time to validate the model prediction accuracy. The results show that, the proposed model has better performance than the traditional bidirectional LSTM model. The improved seagull algorithm can achieve a better optimization effect on the bidirectional LSTM-Attention model. Compared with the existing model and seagull algorithm (SOA) optimized bidirectional LSTM-Attention model, the mean absolute percentage error was reduced by 5.96%, the root mean square error was reduced by 9.87%, and the mean absolute error was reduced by 7.99% in the ISOA-BiLSTM- Attention for bus arrival time prediction. Moreover, the ISOA-BiLSTM-Attention has the largest model decision coefficient R2 value, which indicates the good generalization ability and stability of the proposed model, and can provide good fitness of accuracy for bus arrival time.

Key words: urban traffic, bus arrival time prediction, improved seagull optimization algorithm, bidirectional LSTM model, Attention mechanism

中图分类号:

U121

张兵, 周丹丹, 孙健, 倪训友. 基于双向长短期记忆网络的公交到站时间预测模型[J]. 交通运输系统工程与信息, 2023, 23(2): 148-160.

ZHANG Bing, ZHOU Dan-dan, SUN Jian, NI Xun-you. Bus Arrival Time Prediction Model Based on Bidirectional Long Short-term Memory Network[J]. Journal of Transportation Systems Engineering and Information Technology, 2023, 23(2): 148-160.

导出引用管理器 EndNote|Ris|BibTeX

链接本文: http://www.tseit.org.cn/CN/10.16097/j.cnki.1009-6744.2023.02.016

http://www.tseit.org.cn/CN/Y2023/V23/I2/148

参考文献

[1] 任远, 吕永波, 马继辉, 等. 基于粒子滤波的公交车辆到站时间预测研究[J]. 交通运输系统工程与信息, 2016, 16(6): 142-146. [REN Y, LV Y B, MA J H, et al. Bus arrival time prediction based on particle filter[J]. Journal of Transportation Systems Engineering and Information Technology, 2016, 16(6): 142-146.]

[2] SHARMILA R B, VELAGA N R, HOUDHARY P C. Bus arrival time prediction and measure of uncertainties using survival models[J]. Iet Intelligent Transport Systems, 2020, 14(8): 900-907.

[3] 赖永炫, 杨旭, 曹琦, 等. 一种基于Gradient Boosting 的公交车运行时长预测方法[J]. 大数据, 2019, 5(5): 58- 78. [LAI Y X, YANG X, CAO Q, et al. A bus running length prediction method based on Gradient Boosting [J]. Big Data Research, 2019, 5(5): 58-78.]

[4] PANG J, HUANG J, DU Y, et al. Learning to predict bus arrival time from heterogeneous measurements via recurrent neural network[J]. IEEE Transactions on Intelligent Transportation Systems, 2019, 9(20): 3283- 3293.

[5] CHEN S K, SUN D. A multistate-based travel me schedule model for fixed transit route[J]. Transportation Letters-the International Journal of Transportation Research, 2019, 11(1): 33-42.

[6] 季彦婕, 陆佳炜, 陈晓实, 等. 基于粒子群小波神经网络的公交到站时间预测[J]. 交通运输系统工程与信息, 2016, 16(3): 60-66. [JI Y J, LU J W, CHEN X S, et al. Prediction model of bus arrival Time based on particle swarm optimization and wavelet neural network [J]. Journal of Transportation Systems Engineering and Information Technology, 2016, 16(3): 60-66.]

[7] KUMAR B A, VANAJAKSHI L, SUBRAMANIAN S C. Bus travel time prediction using a time-space discretization approach[J]. Transportation Research Part C: Emerging Technologies, 2017, 79: 308-332.

[8] 安宇航, 马继辉, 刘慧勇, 等. 基于深度学习和粒子滤波的公交车到站时间预测 [J]. 北京交通大学学报, 2022, 46(1): 88-97. [ AN Y H, MA J H, LIU H Y, et al. Prediction of bus arrival time based on deep learning and particle filter[J]. Journal of Beijing Jiaotong University, 2022, 46(1): 88-97.]

[9] 张孝梅, 陈旭梅, 张溪. 基于非参数回归-粒子滤波模型的公交到站时间预测 [J]. 公路交通科技, 2020, 37 (4): 118-124. [ZHANG X M, CHEN X M, ZHANG X. Prediction of bus arrival time based on nonparametric regression and particle filtering model[J]. Journal of Highway and Transportation Research and Development, 2020, 37(4): 118-124.]

[10] XIE Z Y, HE Y R, CHEN C C, et al. Multistep prediction of bus arrival time with the recurrent neural network[J]. Mathematical Problems in Engineering, 2021(2021): 1- 14.

[11] 苗旭, 王忠宇, 吴兵, 等. 考虑前序路段状态的公交到站时间双层BPNN预测模型 [J]. 交通运输系统工程与信息, 2020, 20(2): 127-133. [MIAO X, WANG Z Y, WU B, et al. Bi-layer BPNN prediction model for bus arrival time considering preceding segment state[J]. Journal of Transportation Systems Engineering and Information Technology, 2020, 20(2): 127-133.]

[12] 邝先验, 罗会超, 钟蕊, 等. 基于天牛须小波神经网络的公交到站时间预测 [J]. 吉林大学学报(工学版), 2022, 52(1): 110-117. [KUANG X Y, LUO H C, ZHONG R, et al. Bus arrival time prediction based on wavelet neural network optimized by beetle antennae search[J]. Journal of Jilin University (Engineering and Technology Edition), 2022, 52(1): 110-117.]

[13] LAI Y, EASA S, SUN D, et al. Bus arrival time prediction using wavelet neural network trained by improved particle swarm optimization[J]. Journal of Advanced Transportation, 2020(2020): 1-9.

[14] CUI Z, KE R, PU Z, et al. Stacked bidirectional and unidirectional LSTM recurrent neural network for forecasting network-wide traffic state with missing values [J]. Transportation Research Part C: Emerging Technologies, 2020(118): 1-14.

[15] 王金玉, 金宏哲, 王海生, 等. ISSA 优化 Attention 双向LSTM的短期电力负荷预测 [J]. 电力系统及其自动化学报, 2022, 34(5): 111-117. [WANG J Y, JIN H Z, WANG H S, et al. Short-term power load prediction of bidirectional LSTM with ISSA optimization attention mechanism [J]. Proceedings of the CSU-EPSA, 2022, 34 (5): 111-117.]

基于双向长短期记忆网络的公交到站时间预测模型

Bus Arrival Time Prediction Model Based on Bidirectional Long Short-term Memory Network

PDF

可视化

摘要/Abstract

引用本文

使用本文

参考文献

相关文章 15

编辑推荐

Metrics

[1]	周雨阳, 王沛钰, 陈艳艳. 考虑自我呈现与环保意识的出行方式选择行为异质性分析[J]. 交通运输系统工程与信息, 2023, 23(3): 30-38.
[2]	陈坚, 李兆康, 周涛, 刘柯良, 孙琴梅. 基于用户均衡的双模式接驳公交一体化布局优化[J]. 交通运输系统工程与信息, 2023, 23(3): 56-65.
[3]	谢东繁, 罗钰超, 周广京, 余亚鹏, 王永兴, 赵小梅. 考虑电池退化的单线路电动公交全生命周期运营规划优化策略[J]. 交通运输系统工程与信息, 2023, 23(3): 66-75.
[4]	杜鹏, 何思乐. 基于空间杜宾模型的城市轨道交通车站周边商业地产价格分析[J]. 交通运输系统工程与信息, 2023, 23(3): 94-100.
[5]	苏跃江, 温惠英, 袁敏贤, 吴德馨, 漆巍巍. 基于集成学习和居民属性数据的出行方式预测模型[J]. 交通运输系统工程与信息, 2023, 23(3): 153-160.
[6]	贺鹏, 李雯茜, 李妍, 许奇. 建成环境对轨道交通客流非线性影响的空间特征[J]. 交通运输系统工程与信息, 2023, 23(3): 187-194.
[7]	赵雪亭, 胡立伟. 城市交通拥塞因子时空变化特征及源解析研究[J]. 交通运输系统工程与信息, 2023, 23(3): 300-310.
[8]	陈坚, 张弛, 傅志妍, 刘柯良. 社交网络信息对出行行为的影响研究综述[J]. 交通运输系统工程与信息, 2023, 23(2): 1-10.
[9]	李文翔, 程佳楠, 刘向龙, 穆凯, 蔡近近, 刘魏巍. 出行即服务环境下个体出行链碳足迹监测与评估[J]. 交通运输系统工程与信息, 2023, 23(2): 22-31.
[10]	金盛, 苏弘扬, 张静. 融合出行拓扑与序列分析的车辆时空出行模式挖掘[J]. 交通运输系统工程与信息, 2023, 23(2): 40-53.
[11]	赵雪亭, 胡立伟. 基于典型耦合优化算法的城市交通拥塞点源反演识别研究[J]. 交通运输系统工程与信息, 2023, 23(2): 74-83.
[12]	姚雅慧, 张戎. 基于轨迹数据的货车停车目的识别方法[J]. 交通运输系统工程与信息, 2023, 23(2): 92-99.
[13]	李想, 晏启鹏, 骆晨. 空间异质性建成环境对地铁与公交换乘客流的影响[J]. 交通运输系统工程与信息, 2023, 23(2): 100-110.
[14]	孙会君, 冯宗旭, 郑汉坤. 考虑地铁运营中断下多站协同的接驳公交优化研究[J]. 交通运输系统工程与信息, 2023, 23(2): 111-120.
[15]	刘翔, 陈小鸿, 田茗舒. 成长性视角下建成环境对轨道交通站点客流影响分析[J]. 交通运输系统工程与信息, 2023, 23(2): 121-127.