基于二次分解和融合多特征的短时交通流量组合预测模型

doi:10.16088/j.issn.1001-6600.2022062803

摘要/Abstract

摘要： 考虑到交通流的随机性和非线性特征导致预测精度低的问题,本文提出一种基于二次分解和融合多特征的组合预测模型。利用时序分解方法将提取交通流量中的趋势性和周期性特征,通过优化后的变分模态分解对残差分量进行二次分解,并对所得分量进行重构;使用相关系数法选取交通流的外部特征,建立3个相异模型对融合外部特征后的分量进行预测;利用强化学习优化各模型的权重,加权求和得到最终的预测结果。利用长沙市区的交通流量进行仿真分析,结果表明:与长短时记忆神经网络模型、卷积神经网络和门控循环单元的组合模型、二次分解后的BP和二次分解后的轻量级梯度提升机相比,本文建立的模型对城市道路交通流的预测效果更好,平均绝对误差为2.622,均方根误差为3.479,均优于对比模型的预测误差,验证了模型的有效性。

关键词: 短时交通流量预测, 时序分解, 特征选择, Q-Learning, 组合模型

Abstract: Considering the random and nonlinear characteristics of traffic flow that result in low prediction accuracy, a combined prediction model based on quadratic decomposition and fusion of multiple features is proposed, where the trend and periodic features in traffic flow are extracted by using the time series decomposition method, the residual components are quadratically decomposed by the optimized variational mode decomposition, the scored amount is reconstructed, the external features of the traffic flow are selected by the correlation coefficient method, and three different models are established to predict the components after fusing the external features. Reinforcement learning is used to optimize the weights of each model, and the final prediction result is obtained by weighted sum. Using the simulation analysis of traffic flow in Changsha urban area, experimental results show that compared with the long short-term memory neural network model. Combined model of convolutional neural networks and gated cyclic units, BP after quadratic decomposition and lightweight gradient lifter after quadratic decomposition, the model established in this paper has a better prediction effect on urban road traffic flow, with an average absolute error of 2.622 and a root mean square error of 3.479. The prediction errors are better than the existing models, which verifies the effectiveness of the proposed model.

Key words: short-time traffic forecasting, time series decomposition, feature selection, Q-Learning, combined model

中图分类号: U491.14

陈昆, 曲大义, 王少杰, 王其坤. 基于二次分解和融合多特征的短时交通流量组合预测模型[J]. 广西师范大学学报（自然科学版）, 2023, 41(4): 33-46.

CHEN Kun, QU Dayi, WANG Shaojie, WANG Qikun. Short-Time Traffic Flow Combination Prediction Model Based on Quadratic Decomposition and Fusion of Multiple Features[J]. Journal of Guangxi Normal University(Natural Science Edition), 2023, 41(4): 33-46.

参考文献

[1] SUN Z Y, HU Y J, LI W, et al. Prediction model for short-term traffic flow based on a K-means-gated recurrent unit combination
[J]. IET Intelligent Transport Systems, 2022, 16(5): 675-690. DOI: 10.1049/itr2.12165.
[2] WILLIAMS B M, HOEL L A. Modeling and forecasting vehicular traffic flow as a seasonal ARIMA process: theoretical basis and empirical results[J]. Journal of Transportation Engineering, 2003, 129(6): 664-672. DOI: 10.1061/(ASCE)0733-947X(2003)129:6(664).
[3] 黑凯先, 曲大义, 周警春, 等. 基于随机森林决策树的行驶车辆换道行为识别[J].青岛理工大学报, 2020, 41(1): 115-120. DOI: 10.3969/j.issn.1673-4602.2020.01.018.
[4] 邹宗民, 郝龙, 李全杰, 等. 基于粒子群优化-支持向量回归的高速公路短时交通流预测[J].科学技术与工程, 2021, 21(12):5118-5123. DOI: 10.3969/j.issn.1671-1815.2021.12.054.
[5] 杨正理, 陈海霞, 王长鹏, 等. 大数据背景下城市短时交通流预测[J].公路交通科技, 2019, 36(2):136-143. DOI: 10.3969/j.issn.1002-0268.2019.02.018.
[6] 许伦辉, 陈凯勋. 基于改进萤火虫算法优化BP神经网络的路网速度分布预测[J].广西师范大学学报(自然科学版), 2019, 37(2):1-8.DOI: 10.16088/j.issn.1001-6600.2019.02.001.
[7] 陈明猜, 於东军, 戚湧. 基于FOA-RBF网络的城市道路短时交通流预测[J].南京邮电大学学报(自然科学版), 2018, 38(2):103-110. DOI: 10.14132/j.cnki.1673-5439.2018.02.017.
[8] 宋瑞蓉, 王斌君, 仝鑫, 等. 基于改进果蝇的混合小波神经网络交通流预测[J].科学技术与工程, 2021, 21(15):6394-6401. DOI: 10.3969/j.issn.1671-1815.2021.15.040.
[9] ZHENG J, HUANG M. Traffic flow forecast through time series analysis based on deep learning[J]. IEEE Access, 2020, 8: 82562-82570. DOI: 10.1109/ACCESS.2020.2990738.
[10] 王博文, 王景升, 朱茵, 等. 基于ARMA-SVR的短时交通流量预测模型研究[J].公路交通科技, 2021, 38(11):126-133. DOI: 10.3969/j.issn.1002-0268.2021.11.015.
[11] MA T, ANTONIOU C, TOLEDO T. Hybrid machine learning algorithm and statistical time series model for network-wide traffic forecast[J]. Transportation Research Part C: Emerging Technologies, 2020, 111: 352-372. DOI: 10.1016/j.trc.2019.12.022.
[12] 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. DOI: 10.16097/j.cnki.1009-6744.2018.01.012.
[13] 卢生巧, 黄中祥. 基于深度学习的短时交通流预测模型[J].交通科学与工程, 2020, 36(3):74-80. DOI: 10.16544/j.cnki.cn43-1494/u.2020.03.012.
[14] 邵春福, 薛松, 董春娇, 等. 考虑时空相关性的网络交通流短期预测[J].北京交通大学学报, 2021, 45(4):37-43. DOI: 10.11860/j.issn.1673-0291.20200115.
[15] 许伦辉, 游黄阳. 基于特性和影响因素分析的短时交通流预测[J].广西师范大学学报(自然科学版), 2013, 31(1):1-5. DOI: 10.16088/j.issn.1001-6600.2013.01.001.
[16] 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. DOI: 10.16182/j.issn1004731x.joss.19-0565.
[17] 李桃迎, 王婷, 张羽琪. 考虑多特征的高速公路交通流预测模型[J].交通运输系统工程与信息, 2021, 21(3):101-111. DOI: 10.16097/j.cnki.1009-6744.2021.03.013.
[18] 殷礼胜, 孙双晨, 魏帅康, 等. 基于自适应VMD-Attention-BiLSTM的交通流组合预测模型[J].电子测量与仪器学报, 2021, 35(7):130-139. DOI: 10.13382/j.jemi.B2003575.
[19] ZHAO J D, YU Z X, YANG X, et al. Short term traffic flow prediction of expressway service area based on STL-OMS[J]. Physica A: Statistical Mechanics and Its Applications, 2022, 595: 126937-126952. DOI: 10.1016/j.physa.2022.126937.
[20] ZHANG Y R, ZHANG Y L, HAGHANI A. A hybrid short-term traffic flow forecasting method based on spectral analysis and statistical volatility model[J]. Transportation Research Part C: Emerging Technologies, 2014, 43: 65-78. DOI: 10.1016/j.trc.2013.11.011.
[21] 吴玲玲, 尹莉莉, 任其亮. 一种EMD和DE-BPNN组合优化的短时交通流预测方法[J].重庆理工大学学报(自然科学), 2021, 35(12):155-163. DOI: 10.3969/j.issn.1674-8425(z).2021.12.020.
[22] 彭延峰, 彭志华, 刘燕飞. 基于ASNBD-CMFE特征信息提取的短时交通流预测[J].北京交通大学学报, 2021, 45(2):28-35. DOI: 10.11860/j.issn.1673-0291.20200082.
[23] LEMPEL A, ZIV J. On the complexity of finite sequences[J]. IEEE Transactions on Information Theory, 1976, 22(1): 75-81. DOI: 10.1109/TIT.1976.1055501.
[24] PASCANU R, MIKOLOV T, BENGIO Y. On the difficulty of training recurrent neural networks[C]// Proceedings of the 30th International Conference on Machine Learning. New York: PMLR, 2013, 28(3): 1310-1318.
[25] 王扬, 陈智斌, 吴兆蕊, 等. 强化学习求解组合最优化问题的研究综述[J].计算机科学与探索, 2022, 16(2):261-279. DOI: 10.3778/j.issn.1673-9418.2107040.

Metrics

Viewed

Full text

Abstract

Cited

Shared

Discussed