A Short-term Traffic Flow Prediction Method Based on Time Series Data Decomposition and Reconstruction
-
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。Abstract: 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 decomposition reconstruction is constructed by combining the parameter optimization based variational mode decomposition (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 determine 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 deterministic 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 values of each reconstructed component are nonlinearly integrated using the BIGRU prediction model to obtain the final prediction results. The measured data of the flow of Shanghai North-South Expressway and California Expressway network are used for validation, The results show that in the NBDX08(1) dataset, the corresponding mean absolute 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 reconstruction 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.
-
图 2 基于OCCSA算法优化的VMD分解流程
Figure 2. VMD decomposition flow process based on OCCSA optimization
图 5 基于时序数据分解重构的模型预测流程
Figure 5. A model prediction process based on time-series data decomposition and reconstruction
图 20 不同检测器数据的5种模型预测效果
Figure 20. Predictive effects of five models for different detector data
表 1 各IMF的τ，m，ε值
Table 1. τ, m, ε for each IMF
参数 IMF 1 2 3 4 5 6 7 8 9 10 m 2 2 2 2 2 2 2 2 2 5 τ 7 13 7 7 6 6 10 6 5 3 ε 1.0366 8.5314 2.9312 1.2579 0.9831 0.7574 0.7734 0.7740 0.5766 0.5743 
下载: 导出CSV
表 2 各固有模态分量的DET和LAM
Table 2. DET and LAM for each intrinsic modal component
指标 IMF 1 2 3 4 5 6 7 8 9 10 DET 1 1 1 0.9752 0.8796 0.6648 0.5500 0.5343 0.5905 0.9775 LAM 1 1 1 0.9893 0.9421 0.8110 0.7387 0.6326 0.4630 0.3761
下载: 导出CSV
表 3 模型参数设置说明
Table 3. Explanation of model parameter settings
模型 参数设置 SVM 惩罚系数λ=30，核函数宽度参数γ=2 BP神经网络 隐藏层数设为1，隐藏层神经元个数为50，激活函数为Sigmoid函数。学习率设为0.9 BILSTM 3层网络结构，隐藏层数为1，隐藏层神经元个数为128，学习率设为0.005，批量处理大小为32，迭代次数为200 BIGRU 3层网络结构，隐藏层数设为1，隐藏层神经元个数设为128，学习率设为0.001，批量处理大小为32，迭代次数为150 EEMD-BIGRU EEMD的白噪声幅度值系数设为0.2，集成次数设为500，BIGRU参数设置同上 VMD-BIGRU VMD参数通过观察中心频率确定k=5，依经验设置惩罚因子α=2000，BIGRU参数设置同上 VMD-RQA-BIGRU VMD参数通过观察中心频率确定k=5，依经验设置惩罚因子α=2000，BIGRU参数设置同上 OCCSA-VMD-RQA-BIGRU VMD参数通过OCCSA寻优确定k=10，α=1932，BIGRU参数设置同上
下载: 导出CSV
表 4 NBDX08(1)检测器不同模型的预测性能指标对比
Table 4. Comparison of predictive performance metrics for different models of the NBDX08(1)detector
模型 MAE RMSE MAPE/% BP神经网络 9.89 11.17 24.98 SVM 8.02 9.75 23.01 BILSTM 6.94 8.72 20.53 BIGRU 6.37 7.26 19.27 EEMD-BIGRU 5.98 6.42 16.68 VMD-BIGRU 5.21 6.08 14.95 VMD-RQA-BIGRU 4.66 5.73 11.29 OCCSA-VMD-RQA-BIGRU 4.28 5.25 10.59
下载: 导出CSV
表 5 760101号检测器不同模型的预测性能指标对比
Table 5. Comparison of predictive performance metrics for different models of the 760101 detector
模型 MAE RMSE MAPE/% BP神经网络 35.36 45.12 16.28 SVM 34.08 43.53 15.21 BILSTM 31.23 41.07 13.56 BIGRU 30.52 40.75 12.97 EEMD-BIGRU 29.87 40.33 12.59 VMD-BIGRU 24.21 33.26 11.57 VMD-RQA-BIGRU 23.25 31.36 10.06 OCCSA-VMD-RQA-BIGRU 21.57 29.05 9.81
下载: 导出CSV
-
[1] 傅成红, 杨书敏, 张阳. 改进支持向量回归机的短时交通流预测[J]. 交通运输系统工程与信息, 2019, 19 (4): 130-134, 148.FU C H, YANG S M, ZHANG Y. Promoted short-term traffic flow prediction model basedon deep learning and support vector regression[J]. Journal of Transportation Systems Engineering and Information Technology, 2019, 19(4): 130-134, 148. (in Chinese) [2] 姚俊峰, 何瑞, 史童童, 等. 基于机器学习的交通流预测方法综述[J]. 交通运输工程学报, 2023, 23 (3): 44-67.YAO J F, HE R, SHI T T, et al. Review onmachine learning based traffic flow prediction methods[J]. Journal of Traffic and Transportation Engineering, 2023, 23 (3): 44-67. (in Chinese) [3] ASMA B, YOUCEF D, DJENOURI D, et al. A recurrent neural network for urban long-term traffic flow forecasting[J]. Applied Intelligence, 2020, 50 (10): 1-14. [4] 陈喜群, 周凌霄, 曹震. 基于图卷积网络的路网短时交通流预测研究[J]. 交通运输系统工程与信息, 2020, 20 (4): 49-55.CHEN X Q, ZHOU L X, CAO Z. Short-term network widetraffic prediction based on graph convolutional network[J]. Journal of Transportation Systems Engineering and Information Technology, 2020, 20 (4): 49-55. (in Chinese) [5] WANG R J, SHI W Z, LIU X L, et al. An adaptive cutoff frequency selection approach for fast fourier transform method and its application into short-term traffic flow forecasting[J]. ISPRS International Journal of Geo Information, 2020, 9 (12): 731-731. doi: 10.3390/ijgi9120731 [6] LUO B R, CAO R H, YANG W B, et al. Analysing and predicting the fine-scale distribution of traffic particulate matter in urban non-motorized lanes by using wavelet transform and random forest methods[J]. Stochastic Environmental Research and Risk Assessment, 2023, 37 (7): 2657-2676. doi: 10.1007/s00477-023-02411-6 [7] JAVAD S, NAFISEH S. Neuro-fuzzy modeling of data singular spectrum decomposition and traffic flow prediction[J]. Iranian Journal of Science and Technology, Transactions of Electrical Engineering, 2020, 44 (1): 519-535. doi: 10.1007/s40998-019-00227-1 [8] QUANG D T, Q. H T, VAN M N. An enhanced ensemble based long short-term memory approach for traffic volume prediction[J]. Computers, Materials & Continua, 2024, 78 (3): 3585-3602. [9] 沈富鑫, 邴其春, 张伟健, 等. 基于CEEMD-GRU组合模型的快速路短时交通流预测[J]. 河北科技大学学报, 2021, 42 (5): 454-461.SHEN F X, BING Q C, ZHANG W J, et al. Short term traffic flow prediction of express way based on CEEMD-GRU combination model[J]. Journal of Hebei University of Scienceand Technology, 2021, 42 (5): 454-461. (in Chinese) [10] 沈富鑫, 邴其春, 张伟健, 等. 基于CEEMD-AN-ABC-LSTM组合模型的短时交通流预测[J]. 青岛理工大学学报, 2022, 43 (5): 96-103, 119.SHEN F X, BING Q C, ZHANG W J, et al. Short term traffic flow prediction based on CEEMDAN-ABC-LSTM combined model[J]. Journal of Qingdao University of Technology, 2022, 43 (5): 96-103, 119. (in Chinese) [11] 邴其春, 张伟健, 沈富鑫, 等. 基于变分模态分解和LSTM的短时交通流预测[J]. 重庆理工大学学报(自然科学), 2023, 37 (5): 169-177.BING Q C, ZHANG W J, SHEN F X, et al. Short-term traffic flow prediction based on variational modal decomposition and LSTM[J]. Journal of Chongqing University of Technology (Natural Science), 2023, 37 (5): 169-177. (in Chinese) [12] WEI Y, CHEN M C. Forecasting the short-term metro passenger flow with empirical mode decomposition and neural networks[J]. Transportation Research Part C: Emerging Technologies, 2012, 21 (1): 148-162. doi: 10.1016/j.trc.2011.06.009 [13] CHEN X Q, CHEN H X, YANG Y S, et al. Traffic flow prediction by an ensemble framework with data denoising and deep learning model[J]. Physica A: Statistical Mechanics and its Applications, 2021, 565: 125574. doi: 10.1016/j.physa.2020.125574 [14] HUANG H, MAO J N, LU W K, et al. DE-ASeq2Seq: an attention based sequence to sequence model for short-term metro passenger flow prediction within decomposition ensemble strategy[J]. Transportation Research Part C: Emerging Technologies, 2023, 146: 103965. doi: 10.1016/j.trc.2022.103965 [15] 崔文岳, 谷远利, 赵胜利, 等. 基于有向图卷积与门控循环单元的短时交通流预测方法[J]. 交通信息与安全, 2023, 41 (2): 121-128. doi: 10.3963/j.jssn.1674-4861.2023.02.013CUI W Y, GU Y L, ZHAO S L, et al. A method of predicting short-term traffic flows based on a DGC-GRU model[J]. Journal of Transport Information and Safety, 2023, 41(2): 121-128. (in Chinese) doi: 10.3963/j.jssn.1674-4861.2023.02.013 [16] 戢晓峰, 孔晓丽, 陈方, 等. 基于ETC数据和A-BiLSTM神经网络的高速公路节假日短时交通流预测模型[J]. 交通信息与安全, 2023, 41 (3): 166-174. doi: 10.3963/j.jssn.1674-4861.2023.03.018JI X F, KONG X L, CHEN F, et al. A forecasting model of short-term traffic flow on expressways during holidays based on ETC data and A-BiLSTM neural network models[J]. Journal of Transport Information and Safety, 2023, 41(3): 166-174. (in Chinese) doi: 10.3963/j.jssn.1674-4861.2023.03.018 [17] STROZZI F, POZZI R. Improving time seriesfeatures identification by means of convolutionnal neural networks and recurrence plot[J]. IFAC PapersOnLine, 2022, 55 (10): 601-606. doi: 10.1016/j.ifacol.2022.09.464 [18] XUE J K, SHEN B. A novel swarm intelligence optimization approach: sparrow search algorithm[J]. Systems Science Control Engineering, 2020, 8 (1): 22-34. doi: 10.1080/21642583.2019.1708830 [19] MOHAMMAD D, PAVEL T. Osprey optimiz-ation algorithm: a new bio-inspired metaheuristic algorithm for solving engineering optimization problems[J]. Frontiers in Mechanicaling, 2023 (8): 1126450. [20] 李爱莲, 全凌翔, 崔桂梅, 等. 融合正余弦和柯西变异的麻雀搜索算法[J]. 计算机工程与应用, 2022, 58 (3): 91-99.LI A L, QUAN L X, CUI G M, et al. Sparrow search algorithm combining sinecosine and Cauchy mutation[J]. Computer Engineering and Applications, 2022, 58(3): 91-99. (in Chinese) [21] 聂铃, 张剑, 胡茂政. 基于CEEMDAN分解的短时交通流组合预测[J]. 计算机工程与应用, 2022, 58 (11): 279-286. doi: 10.3778/j.issn.1002-8331.2106-0305NIE L, ZHANG J, HU M Z. Short-term traffic flow combination prediction based on CEEMDAN decomposition[J]. Computer Engineering and Applications, 2022, 58(11): 279-286. (in Chinese) doi: 10.3778/j.issn.1002-8331.2106-0305 [22] 邴其春, 龚勃文, 杨兆升, 等. 一种组合核相关向量机的短时交通流局域预测方法[J]. 哈尔滨工业大学学报, 2017, 49 (3): 144-149.BING Q C, GONG B W, YANG Z S, et al. A short-term traffic flow local prediction method of combined kernel function relevance vector machine[J]. Journal of Harbin Institute of Technology, 2017, 49 (3): 144-149. (in Chinese) [23] 李巧茹, 池维源, 陈亮, 等. 基于相空间重构和PSO-GPR的短时交通流预测[J]. 交通信息与安全, 2019, 37 (2): 70-76. doi: 10.3963/j.issn.1674-4861.2019.02.010LI Q R, CHI W Y, CHEN L, et al. Short-term traffic flow forecast based on phase space reconstruction and PSO-GPR[J]. Journal of Transport Information and Safety, 2019, 37 (2) : 70-76. (in Chinese) doi: 10.3963/j.issn.1674-4861.2019.02.010 [24] 许伦辉, 唐德华, 邹娜, 等. 基于非线性时间序列分析的短时交通流特性分析[J]. 重庆交通大学学报(自然科学版), 2010, 29 (1): 110-113.XU L H, TANG D H, ZOU N, et al. Analysis on short-term traffic flow characteristics based on nonlinear time series analysis[J]. Journal of Chongqing Jiaotong University(Natural Science), 2010, 29 (1): 110-113. (in Chinese) [25] XIONG H, SHANG P J, BIAN S H. Detecting intrinsic dynamics of traffic flow with recurrence analysis and empirical mode decomposition[J]. Physica A: Statistical Mechanics and its Applications, 2017, 474: 70-84. doi: 10.1016/j.physa.2017.01.060 -
点击查看大图
图(20) / 表(5)
计量
- 文章访问数: 66
- HTML全文浏览量: 19
- PDF下载量: 12
- 被引次数: 0