基于VAR-LRTC-TNN的交通流量数据补全框架模型

doi:10.3969/j.issn.1674-0696.2024.06.07

摘要/Abstract

摘要： 从各类传感系统收集到的交通流数据往往会因探测器或通信故障等缘故出现数据连续性的缺失，故准确补全缺失的交通流数据对制定合理的交通管理策略至关重要。鉴于交通流数据具有低秩的特性，通过低秩张量补全模型可较好地刻画出交通流数据的全局一致性，但却无法很好地捕捉数据的局部变化趋势，一定程度上影响了效果。基于此，提出了将VAR模型和基于残差序列的LRTC-TNN模型相结合的交通流补全框架模型；采用VAR模型对缺失数据进行粗略估计，移除平均趋势，利用LRTC-TNN模型对残差时间序列进行补全，再将平均趋势还原，从而完成对交通流量数据的高精度补全；该方法不仅保留了交通流数据的全局结构，还考虑了数据局部变化的特征。研究结果表明：与基于原始交通流量数据的填充方法相比，该模型框架对单传感器和多传感器数据的连续性缺失均具有更高的补全精度。

Abstract: The traffic flow data collected from various sensing systems frequently occurs continuously data missing due to detector and communication failure. Therefore, it is crucial to accurately complete the missing traffic flow data to develop a reasonable traffic management strategy. Given the low-rank nature of traffic flow data, the low-rank tensor completion model could effectively characterize the global consistency of traffic flow data, but it failed to well capture the local change trend of the data, which to some extent affected the effectiveness. In view of the above problems, a traffic flow completion framework model by combining VAR model with LRTC-TNN model based on residual sequence was proposed. The VAR model was used to roughly estimate the missing data and remove the average trend. The LRTC-TNN model was used to complete the residual time series and then restore the average trend, so as to achieve high-precision completion of traffic flow data. The proposed method not only retained the global structure of traffic flow data, but also considered the characteristics of local variation of the data. The research results indicate that the proposed model framework has higher completion accuracy for the continuously missing of data for single and multiple sensors than the filling method based on the original traffic flow data does.

孙秋霞1，王淇1，李勍1，孙璐2，贾秀燕1. 基于VAR-LRTC-TNN的交通流量数据补全框架模型[J]. 重庆交通大学学报（自然科学版）, 2024, 43(6): 47-53.

SUN Qiuxia1, WANG Qi1, LI Qing1, SUN Lu2, JIA Xiuyan1. Traffic Flow Data Completion Framework Model Based on VAR-LRTC-TNN[J]. Journal of Chongqing Jiaotong University(Natural Science), 2024, 43(6): 47-53.

参考文献

［1］张阳, 廖晓烨, 杨书敏, 等. 一种结构优化的深度信任网络短时交通流预测［J］. 重庆交通大学学报(自然科学版), 2023, 42(11): 126-133.
ZHANG Yang, LIAO Xiaoye, YANG Shumin, et al. Short-term traffic flow prediction with structure-optimized deep belief network［J］. Journal of Chongqing Jiaotong University(Natural Science), 2023, 42(11): 126-133.
［2］ DENG Lei, LIU Xiaoyang, ZHENG Haifeng, et al. Graph spectral regularized tensor completion for traffic data imputation［J］. IEEE Transactions on Intelligent Transportation Systems, 2022, 23(8): 10996-11010.
［3］ EL FAOUZI N E, LEUNG H, KURIAN A. Data fusion in intelligent transportation systems: progress and challenges: A survey［J］. Information Fusion, 2011, 12(1): 4-10.
［4］ ZHANG Xinyu, ZHANG Yong, WEI Xiulan, et al. Traffic forecasting with missing data via low rank dynamic mode decomposition of tensor［J］. IET Intelligent Transport Systems, 2022, 16(9): 1164-1176.
［5］ LIN Kaitong, ZHENG Haifeng, FENG Xinxin, et al. A novel spatial-temporal regularized tensor completion algorithm for traffic data imputation［C］// 2018 10th International Conference on Wireless Communications and Signal Processing (WCSP). Hangzhou, China. IEEE, 2018: 1-6.
［6］ ZHU Yanmin, LI Zhi, ZHU Hongzi, et al. A compressive sensing approach to urban traffic estimation with probe vehicles［J］. IEEE Transactions on Mobile Computing, 2013, 12(11): 2289-2302.
［7］ QU Li, LI Li, ZHANG Yi, et al. PPCA-based missing data imputation for traffic flow volume: A systematical approach［J］. IEEE Transactions on Intelligent Transportation Systems, 2009, 10(3): 512-522.
［8］ LI Li, LI Yuebiao, LI Zhiheng. Efficient missing data imputing for traffic flow by considering temporal and spatial dependence［J］. Transportation Research Part C: Emerging Technologies, 2013, 34: 108-120.
［9］ YU Jingru, STETTLER M E J, ANGELOUDIS P, et al. Urban network-wide traffic speed estimation with massive ride-sourcing GPS traces［J］. Transportation Research Part C: Emerging Technologies, 2020, 112: 136-152.
［10］ CHEN Xinyu, HE Zhaocheng, SUN Lijun. A Bayesian tensor decomposition approach for spatiotemporal traffic data imputation［J］. Transportation Research Part C: Emerging Technologies, 2019, 98: 73-84.
［11］ TAN Huachun, FENG Guangdong, FENG Jianshuai, et al. A tensor-based method for missing traffic data completion［J］. Transportation Research Part C: Emerging Technologies, 2013, 28: 15-27.
［12］ CARROLL J D, CHANG J J. Analysis of individual differences in multidimensional scaling via an n-way generalization of “Eckart-Young” decomposition［J］. Psychometrika, 1970, 35(3): 283-319.
［13］ TUCKER L R. Implications of factor analysis of three-way matrices for measurement of change［J］. Problems in Measuring Change, 1963, 15(3): 122-137.
［14］ ACAR E, DUNLAVY D M, KOLDA T G, et al. Scalable tensor factorizations for incomplete data［J］. Chemometrics and Intelligent Laboratory Systems, 2011, 106(1): 41-56.
［15］ TAN Huachun, FENG Jianshuai, CHEN Zhengdong, et al. Low multilinear rank approximation of tensors and application in missing traffic data［J］. Advances in Mechanical Engineering, 2014, 6: 157597.
［16］高志军, 刘懿如, 王江锋, 等. 基于梯度下降 Tucker 分解的高速公路数据质量控制算法［J］. 北京交通大学学报, 2019, 43(6): 50-55.
GAO Zhijun, LIU Yiru, WANG Jiangfeng, et al. Data quality control algorithm for freeway based on gradient descent Tucker decomposition［J］. Journal of Beijing Jiaotong University(Natural Science), 2019, 43(6): 50-55.
［17］ LIU Ji, MUSIALSKI P, WONKA P, et al. Tensor completion for estimating missing values in visual data［J］. IEEE Transactions on Pattern Analysis and Machine Intelligence, 2013, 35(1): 208-220.
［18］ HU Yao, ZHANG Debing, YE Jieping, et al. Fast and accurate matrix completion via truncated nuclear norm regularization［J］. IEEE Transactions on Pattern Analysis and Machine Intelligence, 2013, 35(9): 2117-2130.
［19］ GU Shuhang, ZHANG Lei, ZUO Wangmeng, et al. Weighted nuclear norm minimization with application to image denoising［C］// 2014 IEEE Conference on Computer Vision and Pattern Recognition. Columbus, O H, USA. IEEE, 2014: 2862-2869.
［20］ YAO Quanming, KWOK J T, WANG Taifeng, et al. Large-scale low-rank matrix learning with nonconvex regularizers［J］. IEEE Transactions on Pattern Analysis and Machine Intelligence, 2019, 41(11): 2628-2643.
［21］教欣萍, 王江锋, 陈磊, 等. 基于HALRTC理论的短时交通流预测算法［J］. 山东科学, 2019, 32(6): 62-68.
JIAO Xinping, WANG Jiangfeng, CHEN Lei, et al. Short-term traffic flow prediction algorithm based on HALRTC theory［J］. Shandong Science, 2019, 32(6): 62-28.
［22］ HUANG Longting, SO H C, CHEN Yuan, et al. Truncated nuclear norm minimization for tensor completion［C］// 2014 IEEE 8th Sensor Array and Multichannel Signal Processing Workshop (SAM). A Coruna, Spain. IEEE, 2014: 417-420.
［23］ HAN Zifa, LEUNG C S, HUANG Longting, et al. Sparse and truncated nuclear norm-based tensor completion［J］. Neural Processing Letters, 2017, 45(3): 729-743.
［24］ XUE Shengke, QIU Wenyuan, LIU Fan, et al. Low-rank tensor completion by truncated nuclear norm regularization［C］// 2018 24th International Conference on Pattern Recognition (ICPR). Beijing, China. IEEE, 2018: 2600-2605.
［25］ CHEN Xinyu, YANG Jinming, SUN Lijun. A nonconvex low-rank tensor completion model for spatiotemporal traffic data imputation［J］. Transportation Research Part C: Emerging Technologies, 2020, 117: 102673.
［26］ CHEN Xinyu, LEI Mengying, SAUNIER N, et al. Low-rank autoregressive tensor completion for spatiotemporal traffic data imputation［J］. IEEE Transactions on Intelligent Transportation Systems, 2022, 23(8): 12301-12310.
［27］ NIE Tong, QIN Guoyang, SUN Jian. Truncated tensor Schatten p-norm based approach for spatiotemporal traffic data imputation with complicated missing patterns［J］. Transportation Research Part C: Emerging Technologies, 2022, 141: 103737.
［28］ RAN Bin, TAN Huachun, WU Yuankai, et al. Tensor based missing traffic data completion with spatial-temporal correlation［J］. Physica A: Statistical Mechanics and Its Applications, 2016, 446: 54-63.
［29］王涛，谢思红，黎文皓，等．基于FFOS-ELM和PF的短时交通流自适应预测模型［J］．重庆交通大学学报(自然科学版), 2021, 40(6): 21-27.
WANG Tao, XIE Sihong, LI Wenhao, et al. Short-term traffic flow adaptive prediction model based on FFOS-ELM and PF［J］. Journal of Chongqing Jiaotong University(Natural Science), 2021, 40(6): 21-27.
［30］ DISSANAYAKE B, HEMACHANDRA O, LAKSHITHA N, et al. A comparison of ARIMAX, VAR and LSTM on multivariate short-term traffic volume forecasting［C］// Conference of Open Innovations Association, FRUCT. FRUCT Oy, 2021(28): 564-570.
［31］ CHEN Chenyi, WANG Yin, LI Li, et al. The retrieval of intra-day trend and its influence on traffic prediction［J］. Transportation Research Part C: Emerging Technologies, 2012, 22: 103-118.