我国自迈入汽车时代以来，全国各地的机动车数量不断猛增，进而也引发了严重的交通拥堵问题。当前，智能交通系统（Intelligent Transportation System, ITS）被当作是解决道路拥堵的可行方法，而针对短时交通流的预测则是其中重要一环。对交通流量实施精准高效的预测，有助于缓解城市交通拥堵、提高城市交通效率，对推动智慧交通与智慧城市协同发展起着至关重要的作用。如何提升短时交通流预测的精度已经成为众多学者研究的重点，为了提升预测模型的准确率，本文基于道路实测数据，将深度学习、群体智能优化算法以及信号分解技术相结合，对交通流的短时预测开展深入研究，具体内容如下：

1）利用实测数据对交通流进行时间相关性分析，找出与被预测时段强相关的历史数据长度，由此确定出模型输入数据的形式，为后续的研究工作奠定基础。

2）提出了基于SSA-LSTM的短时交通流预测模型。首先，对长短时记忆（Long Short-Term Memory, LSTM）神经网络模型进行仿真，并与反向传播（Back Propagation, BP）神经网络和循环神经网络（Recurrent Neural Network, RNN）对比，验证了LSTM模型的优越性。然后，为了解决LSTM在参数确定上没有特定标准和方法的缺陷，选择引入麻雀搜索算法（Sparrow Search Algorithm, SSA）来对LSTM网络参数进行寻优，同时基于检验函数，使用粒子群优化算法（Particle Swarm Optimization, PSO）和灰狼优化算法（Grey Wolf Optimization, GWO）作为对比算法，验证了SSA良好的寻优精度及速度。最后利用实测数据对SSA-LSTM模型预测效果进行验证，结果证明所提模型在短时交通流预测上具有较好的性能。

3）提出了基于VMD-SSA-LSTM的短时交通流预测模型。首先分别通过简单组合信号及实测数据将变分模态分解（Variational Mode Decomposition, VMD）和经验模态分解（Empirical Mode Decomposition, EMD）的分解效果进行对比，验证了VMD方法在信号分解上的优势。然后为了继续提升预测精度，在SSA-LSTM的基础上引入VMD方法。使用VMD方法对原始数据进行分解，将分解得到的各分量分别输入SSA-LSTM进行预测输出，各分量预测结果相叠加即为最终预测结果。最后基于实测数据对VMD-SSA-LSTM模型预测效果进行验证，结果证明所提模型精度最高，同时具有良好的可移植性。

Since the advent of the automotive era in China, the number of motor vehicles has soared continuously, which has caused increasingly serious traffic congestion problems. Currently, Intelligent Transportation System (ITS) is regarded as a feasible method to solve road congestion, and short-term traffic flow prediction is an essential part of it. Precise and efficient prediction of traffic flow is helpful to alleviate urban traffic jam and improve urban traffic efficiency, which plays a crucial role in promoting the coordinated development of smart transportation and smart city. How to improve the accuracy of short-term traffic flow prediction has become the research emphasis of many scholars. In order to improve the accuracy of prediction model, this thesis combined deep learning, swarm intelligence optimization algorithm and signal decomposition technology based on road measured data to carry out in-depth research on short-term traffic flow prediction. The specific content is as follows:

 

1) The measured data was used to analyze the time correlation of traffic flow, and the length of historical data strongly correlated with the predicted period was sought out. Then the form of input data of the model was determined, which laid the foundation for the subsequent research work

 

2) A short-term traffic flow prediction model based on SSA-LSTM was proposed. Firstly, the Long Short-Term Memory (LSTM) neural network model was simulated and compared with Back Propagation (BP) neural network and Recurrent Neural Network (RNN) to verify the advantages of the LSTM model. Then, in order to overcome the shortcoming that LSTM does not have specific criteria and methods in the selection of parameters, Sparrow Search Algorithm (SSA) was introduced to optimize LSTM network parameters. Based on the test function, Particle Swarm Optimization (PSO) and Grey Wolf Optimization (GWO) were used as a comparison to verify the good optimization accuracy and speed of SSA. Finally, the prediction effect of SSA-LSTM model was verified based on measured data, and the results show that the proposed model has good performance in short-term traffic flow prediction.

 

3) A short-term traffic flow prediction model based on VMD-SSA-LSTM was proposed. First of all, the Decomposition effects of Variational Mode Decomposition (VMD) and Empirical Mode Decomposition (EMD) were compared by simple combination signals and measured data respectively. The advantages of VMD method in signal decomposition were verified. Then, in order to further improve the prediction accuracy, the VMD method was introduced on the basis of SSA-LSTM. The VMD method was used to decompose the original data, and each component obtained by decomposition was inputted into SSA-LSTM for prediction output. The superposition of the prediction results of each component was the final prediction result. Finally, the prediction effect of VMD-SSA-LSTM model was verified based on the measured data. The results showed that the proposed model has the highest accuracy and good portability.

[1] 陆海亭, 张宁, 黄卫等. 短时交通流预测方法研究进展[J]. 交通运输工程与信息学报, 2009, 7(4):8.
[2] Ahmed M S, Cook A R. Analysis of freeway traffic time-series data by using box-jenkins technique[J]. Transportation Research Record, 1979(722): 1-9.
[3] Voort M , Dougherty M , Watson S . Combining Kohonen maps with ARIMA time series models to forecast traffic flow[J]. Transportation Research Part C Emerging Technologies, 1996, 4(5): 307-318.
[4] Shahriari S, Ghasri M, Sisson S A, et al. Ensemble of ARIMA: combining parametric and bootstrapping technique for traffic flow prediction[J]. Transportmetrica A: Transport Science, 2020, 16(3): 1552-1573.
[5] Okutani I, Stephanedes Y J. Dynamic prediction of traffic volume through Kalman filtering theory[J]. Transportation Research Part B Methodological, 1984, 18(1): 1-11.
[6] Guo J, Huang W, Williams B M. Adaptive Kalman filter approach for stochastic short-term traffic flow rate prediction and uncertainty quantification[J]. Transportation Research Part C: Emerging Technologies, 2014, 43: 50-64.
[7] 贺国光, 马寿峰, 李宇. 基于小波分解与重构的交通流短时预测法[J]. 系统工程理论与实践, 2002,22(9):101-106.
[8] 刘驰, 刘伟铭. 基于小波变换的交通流预测及模型优化[J]. 科学技术与工程, 2008,8(21):5858-5862.
[9] Zhang H, Wang X, Cao J, et al. A multivariate short-term traffic flow forecasting method based on wavelet analysis and seasonal time series[J]. Applied Intelligence, 2018, 48(10): 3827-3838.
[10] Disbro J E, Frame M. Traffic flow theory and chaotic behavior[J]. Transportation Research Record Journal of the Transportation Research Board, 1989, 1225: 109-115.
[11] Chen Y, Zhao J, Juan D. Traffic flow forecasting algorithm research based on chaos theory[C]//2006 6th World Congress on Intelligent Control and Automation. IEEE, 2006.
[12] Pushkar A, Hall F L, Acha-Daza J A. Estimation of speeds from single-loop freeway flow and occupancy data using cusp catastrophe theory model[J]. Transportation research record, 1994, 1457: 149.
[13] 张亚平, 张起森. 尖点突变理论在交通流预测中的应用[J]. 系统工程学报, 2000,15(3):5.
[14] Davis G A, Nihan N L. Nonparametric regression and short-term freeway traffic forecasting[J]. Journal of Transportation Engineering, 1991, 117(2): 178-188.
[15] Yang Z S, Wang Y, Guan Q. Short-term traffic flow prediction method based on SVM[J]. Journal of Jilin University (Engineering and Technology Edition), 2006, 36(6): 881-884.
[16] Castro-Neto M, Jeong Y S, Jeong M K, et al. Online-SVR for short-term traffic flow prediction under typical and atypical traffic conditions[J]. Expert Systems with Applications, 2009, 36(3-part-P2): 6164-6173.
[17] Liu Z, Du W, Yan D, et al. Short-term traffic flow forecasting based on combination of k-nearest neighbor and support vector regression[J]. Journal of Highway and Transportation Research and Development (English Edition), 2018, 12(1): 89-96.
[18] Alajali W, Zhou W, Wen S, et al. Intersection traffic prediction using decision tree models[J]. Symmetry, 2018, 10(9): 386.
[19] Huang W, Song G, Hong H, et al. Deep architecture for traffic flow prediction: Deep belief networks with multitask learning[J]. IEEE Transactions on Intelligent Transportation Systems, 2014, 15(5): 2191-2201.
[20] Rui F, Zuo Z, Li L. Using LSTM and GRU neural network methods for traffic flow prediction[C]// 2016 31st Youth Academic Annual Conference of Chinese Association of Automation (YAC). IEEE, 2016.
[21] Zheng Z, Chen W, Wu X, et al. LSTM network: A deep learning approach for short-term traffic forecast[J]. IET Intelligent Transport Systems, 2017, 11(2): 68-75.
[22] Yang B, Sun S, Li J, et al. Traffic flow prediction using LSTM with feature enhancement[J]. Neurocomputing, 2019, 332: 320-327.
[23] Bohan H, Yun B. Traffic flow prediction based on BRNN[C]//2019 IEEE 9th International Conference on Electronics Information and Emergency Communication (ICEIEC). IEEE, 2019.
[24] 罗向龙, 焦琴琴, 牛力瑶等. 基于深度学习的短时交通流预测[J]. 计算机应用研究, 2017,34(1):4.
[25] Guo S, Lin Y, Feng N, et al. Attention based spatial-temporal graph convolutional networks for traffic flow forecasting[C]//Proceedings of the AAAI conference on artificial intelligence. 2019, 33(01): 922-929.
[26] 陆文琦, 芮一康, 冉斌等. 智能网联环境下基于混合深度学习的交通流预测模型[J]. 交通运输系统工程与信息, 2020,20(3):7.
[27] Lu S, Zhang Q, Chen G, et al. A combined method for short-term traffic flow prediction based on recurrent neural network[J]. Alexandria Engineering Journal, 2021, 60(1): 87-94.
[28] Eberhart, Shi Y. Particle swarm optimization: Developments, applications and resources[C]// Congress on Evolutionary Computation. IEEE, 2002.
[29] Gao H, Zhao J, Jia L. Short-term traffic flow forecasting model of Elman neural network based on dissimilation particle swarm optimization[C]//2008 IEEE International Conference on Networking, Sensing and Control. IEEE, 2008: 1305-1309.
[30] Chan K Y, Dillon T, Chang E, et al. Prediction of short-term traffic variables using intelligent swarm-based neural networks[J]. IEEE Transactions on Control Systems Technology, 2012, 21(1): 263-274.
[31] Mirjalili S, Mirjalili S M, Lewis A. Grey wolf optimizer[J]. Advances in engineering software, 2014, 69: 46-61.
[32] Yu W, Liu L, Zhang W. Traffic prediction method based on RBF neural network with improved artificial bee colony algorithm[C]//2015 8th International Conference on Intelligent Networks and Intelligent Systems (ICINIS). IEEE, 2015: 141-144.
[33] Yi Q, Zuo Y, Li T, et al. Forecasting of vessel traffic flow using BPNN based on genetic algorithm optimization[C]//2021 International Wireless Communications and Mobile Computing (IWCMC). IEEE, 2021: 1059-1063.
[34] Xue J, Shen B. A novel swarm intelligence optimization approach: sparrow search algorithm[J]. Systems Science & Control Engineering An Open Access Journal, 2020, 8(1): 22-34.
[35] Yan P, Shang S, Zhang C, et al. Research on the processing of coal mine water source data by optimizing BP neural network algorithm with sparrow search algorithm[J]. Ieee Access, 2021, 9: 108718-108730.
[36] Xu T, Wang Y, Zhang D, et al. Prediction on EMS of UAV's data link based on SSA-optimized dual-channel CNN[J]. IEEE Transactions on Electromagnetic Compatibility, 2022, 64(5): 1346-1356.
[37] Yang X, Zou Y, Tang J, et al. Evaluation of short-term freeway speed prediction based on periodic analysis using statistical models and machine learning models[J]. Journal of advanced transportation, 2020, 2020: 1-16.
[38] Chen L, Zheng L, Yang J, et al. Short-term traffic flow prediction: From the perspective of traffic flow decomposition[J]. Neurocomputing, 2020, 413: 444-456.
[39] Zhang Y, Zhang Y, 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.
[40] Yan Y, Wang B, Sheng Q Z, et al. Modelling the publishing process of big location data using deep learning prediction methods[J]. Electronics, 2020, 9(3): 420.
[41] Jiang X, Adeli H. Wavelet packet‐autocorrelation function method for traffic flow pattern analysis[J]. Computer‐Aided Civil and Infrastructure Engineering, 2004, 19(5): 324-337.
[42] Huang N E, Shen Z, Long S R, et al. The empirical mode decomposition and the Hilbert spectrum for nonlinear and non-stationary time series analysis[J]. Proceedings Mathematical Physical & Engineering Sciences, 1998, 454(1971): 903-995.
[43] 罗向龙, 牛国宏, 潘若禹. 交通流量经验模态分解与神经网络短时预测方法[J]. 计算机工程与应用, 2010,46(26):3.
[44] Wang H, Liu L, Dong S, et al. A novel work zone short-term vehicle-type specific traffic speed prediction model through the hybrid EMD–ARIMA framework[J]. Transportmetrica B: Transport Dynamics, 2016, 4(3): 159-186.
[45] Wu Z, Huang N E. Ensemble empirical mode decomposition: a noise-assisted data analysis method[J]. Advances in adaptive data analysis, 2009, 1(01): 1-41.
[46] Dragomiretskiy K, Zosso D. Variational mode decomposition[J]. IEEE transactions on signal processing, 2013, 62(3): 531-544.
[47] Liang D, Xu J, Li S, et al. Short-term passenger flow prediction of rail transit based on VMD-LSTM neural network combination model[C]// 2020 Chinese Control And Decision Conference (CCDC). 2020.
[48] 卢海鹏, 韩莹, 张凯等. 基于VMD-BiLSTM-BLS模型的短时交通流预测[J]. 计算机系统应用, 2022(005):031.
[49] Zhang Y, Pan G, Chen B, et al. Short-term wind speed prediction model based on GA-ANN improved by VMD[J]. Renewable Energy, 2020, 156.
[50] Wen X, Feng Q, Deo R C, et al. Two-phase extreme learning machines integrated with the complete ensemble empirical mode decomposition with adaptive noise algorithm for multi-scale runoff prediction problems[J]. Journal of hydrology, 2019, 570: 167-184.
[51] Zhang G, Liu H, Li P, et al. Load prediction based on hybrid model of VMD-mRMR-BPNN-LSSVM[J]. Complexity, 2020, 2020(4): 1-20.
[52] Beckmann M J. On the theory of traffic flow in networks[J]. Traffic Quarterly, 1967, 21(1).
[53] Boukerche A, Tao Y, Sun P. Artificial intelligence-based vehicular traffic flow prediction methods for supporting intelligent transportation systems[J]. Computer networks, 2020, 182: 107484.
[54] Lv M, Hong Z, Chen L, et al. Temporal multi-graph convolutional network for traffic flow prediction[J]. IEEE Transactions on Intelligent Transportation Systems, 2020, PP(99): 1-12.
[55] 薛薇. SPSS统计分析方法及应用[M]. 电子工业出版社, 2009.
[56] Hopfield, J. J. Neural networks and physical systems with emergent collective computational abilities[J]. Proceedings of the National Academy of Sciences of the United States of America, 1982, 79(8): 2554-2558.
[57] Hochreiter S, Schmidhuber J. Long short-term memory[J]. Neural computation, 1997, 9(8): 1735-1780.
[58] Kumaravel S, Ponnusamy V. An efficient hybrid technique for power flow management in smart grid with renewable energy resources[J]. Energy Sources Part A Recovery Utilization and Environmental Effects, 2020: 1-21.
[59] Yuan J, Zhao Z, Liu Y, et al. DMPPT control of photovoltaic microgrid based on Improved sparrow search algorithm[J]. IEEE Access, 2021, 9: 16623-16629.
[60] Jia J, Yuan S, Shi Y, et al. Improved sparrow search algorithm optimization deep extreme learning machine for lithium-ion battery state-of-health prediction[J]. Iscience, 2022, 25(4): 103-118.
[61] Huang D J, Zhao J P, Su J L. Practical implementation of Hilbert-Huang transform algorithm[J]. Acta Oceanologica Sinica -English Edition-, 2003, 22(1): 1-14.
[62] Dragomiretskiy K, Zosso D. Variational mode decomposition[J]. IEEE Transactions on Signal Processing, 2014, 62(3): 531-544.