基于时序数据分解重构的短时交通流预测方法  

A Short-term Traffic Flow Prediction Method Based on Time Series Data Decomposition and Reconstruction

作　　者：邴其春 赵盼盼 任参政 王雪倩 赵一鸣 BING Qichun;ZHAO Panpan;REN Canzheng;WANG Xueqian;ZHAO Yiming(School of Civil Engineering,Qingdao University of Technology,Qingdao 266520,Shandong,China)

机构地区：[1]青岛理工大学土木工程学院,山东青岛266520

出　　处：《交通信息与安全》2024年第6期112-122,共11页Journal of Transport Information and Safety

基　　金：国家自然科学基金项目(52272311);山东省重点研发计划项目(2019GGX101038)资助。

摘　　要：为了从短时交通流数据中提取蕴含丰富信息的特征分量,进一步提升预测精度,将基于参数优化的变分模态分解(variational mode decomposition,VMD)、递归量化分析(recurrence quantification analysis,RQA)和双向门控循环单元(bidirectional gated recurrent unit,BIGRU)模型相组合,构建了1种基于时序数据分解重构的短时交通流预测方法。采用融合鱼鹰和柯西变异的麻雀优化算法(osprey cauchy sparrow search algorithm,OCSSA)确定变分模态分解的的模态分量个数k和惩罚因子α,获得k个相对平稳的固有模态分量;通过递归量化分析将分解后的模态分量重构为确定性分量、波动分量和趋势分量;在此基础上,针对各重构分量分别构建BIGRU预测模型,并利用BIGRU模型将各重构分量预测结果进行非线性集成,得到最终的预测结果。采用上海市南北快速路和加州高速路网流量实测数据进行实例验证,结果表明:在NBDX08(1)数据集中,相对应的平均绝对误差、均方根误差和平均绝对百分比误差较其他模型平均降低了29.1%,24.5%,46.1%;在760101号数据集中,误差平均降低了19.05%,19.69%,16.46%,验证了本文方法对不同分量进行分解重构可以较为准确的划分和学习交通流分量的特征,在控制模型计算复杂度的同时进一步提升了预测精度。In order to extract signal components with rich feature information from short-term traffic flow data and further improve the prediction accuracy,a short-term traffic flow prediction method based on temporal data decom-position reconstruction is constructed by combining the parameter optimization based variational mode decomposi-tion(VMD),recurrence quantification analysis(RQA),and bidirectional gated recurrent unit(BIGRU)models.The osprey cauchy sparrow search algorithm(OCSSA),which integrates the osprey and cauchy variants,is used to deter-mine the number of modal components and the penalty factor of the variational modal decomposition,and to obtain the relatively smooth intrinsic modal components.The decomposed modal components are reconstructed into the de-terministic components,fluctuating components,and trend components through the recursive quantitative analysis.Based on this,for each reconstructed component the BIGRU prediction model is constructed,and the predicted val-ues of each reconstructed component are nonlinearly integrated using the BIGRU prediction model to obtain the fi-nal prediction results.The measured data of the flow of Shanghai North-South Expressway and California Express-way network are used for validation,The results show that in the NBDX08(1)dataset,the corresponding mean abso-lute error,root-mean-square error,and mean absolute percentage error are reduced by 29.1%,24.5%,and 46.1%on average,respectively,compared with the other models;and the errors in the dataset of No.760101 are reduced by 19.05%,19.69%,and 16.46%on average.These verify that the proposed method for the decomposition and recon-struction of different components can accurately capture and learn the characteristics of traffic flow components,which further improves the prediction accuracy while controlling the computational complexity of the model.

关 键 词：交通运输规划 短时交通流预测 双向门控循环单元 变分模态分解 递归量化分析 

分 类 号：U491.14[交通运输工程—交通运输规划与管理]