A Time Series Decomposition and Reinforcement Learning Ensemble Method for Short-Term Passenger Flow Prediction in Urban Rail Transit

Abstract

Short-term passenger flow prediction (STPFP) helps ease traffic congestion and optimize the allocation of rail transit resources. However, the nonlinear and nonstationary nature of passenger flow time series challenges STPFP. To address this issue, a hybrid model based on time series decomposition and reinforcement learning ensemble strategies is proposed. Firstly, the improved arithmetic optimization algorithm is constructed by adding sine chaotic mapping, a new dynamic boundary strategy, and adaptive T distribution mutations for optimizing variational mode decomposition (VMD) parameters. Then, the original passenger flow data containing nonlinear and nonstationary irregular changes of noise is decomposed into several intrinsic mode functions (IMFs) by using the optimized VMD technology, which reduces the time-varying complexity of passenger flow time series and improves predictability. Meanwhile, the IMFs are divided into different frequency series by fluctuation-based dispersion entropy, and diverse models are utilized to predict different frequency series. Finally, to avoid the cumulative error caused by the direct superposition of each IMF’s prediction result, reinforcement learning is adopted to ensemble the multiple models to acquire the multistep passenger flow prediction result. Experiments on four subway station passenger flow datasets proved that the prediction performance of the proposed method was better than all benchmark models. The excellent prediction effect of the proposed model has important guiding significance for evaluating the operation status of urban rail transit systems and improving the level of passenger service.

Keywords

Variational mode decomposition / Reinforcement learning / Ensemble strategy / Urban rail transit / Short-term passenger flow prediction

References

[1] Zhang JP, Wang FY, Wang KF, Lin WH, Xu X, Chen C. Data-driven intelligent transportation systems: a survey. IEEE Trans Intell Transp Syst, 2011, 12(4): 1624-1639

[2] Ding C, Duan JX, Zhang YR, Wu XK, Yu GZ. Using an ARIMA-GARCH modeling approach to improve subway short-term ridership forecasting accounting for dynamic volatility. IEEE Trans Intell Transp Syst, 2018, 19(4): 1054-1064

[3] Yu L, Feng T, Li T, Cheng L. Demand prediction and optimal allocation of shared bikes around urban rail transit stations. Urban Rail Transit, 2023, 9(1): 57-71

[4] Bezuglov A, Comert G. Short-term freeway traffic parameter prediction: application of grey system theory models. Expert Syst Appl, 2016, 62: 284-292

[5] Smith BL, Demetsky MJ. Traffic flow forecasting: comparison of modeling approaches. J Transp Eng, 1997, 123(4): 261-266

[6] Jiao PP, Li RM, Sun T, Hou ZH, Ibrahim A. Three revised Kalman filtering models for short-term rail transit passenger flow prediction. Math Probl Eng, 2016, 2016: 9717582

[7] Ding C, Wang DG, Ma XL, Li HY (2017) Predicting short-term subway ridership and prioritizing its influential factors using gradient boosting decision trees. Sustainability 8(11):1100

[8] Yan H, Fu LY, Qi Y, Yu DJ, Ye QL. Robust ensemble method for short-term traffic flow prediction. Future Gener Comput Syst, 2022, 133: 359-410

[9] Liu RJ, Wang YH, Zhou H, Qian ZQ. Short-term passenger flow prediction based on wavelet transform and kernel extreme learning machine. IEEE Access, 2019, 7: 158025-158034

[10] Li HY, Wang YT, Xu XY, Qin LQ, Zhang HY. Short-term passenger flow prediction under passenger flow control using a dynamic radial basis function network. Appl Soft Comput, 2019, 83

[11] Huang WH, Song GJ, Hong HK, Xie KQ (2014) Deep architecture for traffic flow prediction: Deep belief networks with multitask learning. IEEE Trans Intell Transp Syst 15(5):2191–2201

[12] Lv YS, Duan YJ, Kang WW, Li ZX, Wang FY (2014) Traffic flow prediction with big data: a deep learning approach. IEEE Trans Intell Transp Syst 16(2):865–873

[13] Liu DQ, Wu Z, Sun SR. Study on subway passenger flow prediction based on deep recurrent neural network. Multimed Tools Appl, 2022, 81(14): 18979-18992

[14] Liu JX, Jiang R, Zhu D, Zhao JD. Short-term subway inbound passenger flow prediction based on AFC Data and PSO-LSTM optimized model. Urban Rail Transit, 2022, 8(1): 56-66

[15] Ma XL, Zhang JY, Du BW, Ding C, Sun LL (2018) Parallel architecture of convolutional bi-directional LSTM neural networks for network-wide metro ridership prediction. IEEE Trans Intell Transp Syst 20(6):2278–2288

[16] Wu JX, Li XW, He DQ, Li Q, Xiang WB (2023) Learning spatial-temporal dynamics and interactivity for short-term passenger flow prediction in urban rail transit. Appl Intell 53(16):19785–19806

[17] Cai LR, Lei MQ, Zhang SY, Yu YD, Zhou T, Qin J (2020) A noise-immune LSTM network for short-term traffic flow forecasting. Chaos 30(2):023135

[18] Chen XQ, Lu JQ, Zhao JS, Qu ZJ, Yan YS, Xian JF (2020) Traffic flow prediction at varied time scales via ensemble empirical mode decomposition and artificial neural network. Sustainability 12(9):3678

[19] Wei Y, Chen MC. Forecasting the short-term metro passenger flow with empirical mode decomposition and neural networks. Transp Res Part C Emerg Technol, 2012, 21(1): 148-162

[20] Yang HF, Chen YPP. Hybrid deep learning and empirical mode decomposition model for time series applications. Expert Syst Appl, 2019, 120: 128-138

[21] Li YP, Ma CX. Short-time bus route passenger flow prediction based on a secondary decomposition integration method. J Transp Eng Part A Syst, 2023, 149: 04022132

[22] Zhang SC, Zhou LX, Chen XQ, Zhang L, Li L, Li M (2020) Network-wide traffic speed forecasting: 3D convolutional neural network with ensemble empirical mode decomposition. Comput Aided Civ Infrastruct Eng 35(10):1132–1147

[23] Liu J, Wu NQ, Qiao Y, Li ZW (2022) Short-term traffic flow forecasting using ensemble approach based on deep belief networks. IEEE Trans Intell Transp Syst 23(1):404–417

[24] Jia XL, Zhou WX, Yang HZ, Li SQ, Chen XP (2023) Short‐term traffic travel time forecasting using ensemble approach based on long short‐term memory networks. IET Intell Transp Syst 17(6):1262–1273

[25] Huang H, Mao JN, Lu WK, Hu GJ, Liu L (2023) DEASeq2Seq: An attention based sequence to sequence model for short-term metro passenger flow prediction within decomposition-ensemble strategy. Transp Res Part C Emerg Technol 146:103965

[26] Diao ZL, Zhang DF, Wang X, Xie K, He SY, Lu X, Li YB (2018) A hybrid model for short-term traffic volume prediction in massive transportation systems. IEEE Trans Intell Transp Syst 20(3):935–946

[27] Liu Y, Song YL, Zhang Y, Liao ZF (2022) WT-2DCNN: A convolutional neural network traffic flow prediction model based on wavelet reconstruction. Physica A 603:127817

[28] Yang X, Xue QC, Yang XX, Yin HD, Qu YC, Li X, Wu JJ (2021) A novel prediction model for the inbound passenger flow of urban rail transit. Inf Sci 566:347–363

[29] Dragomiretskiy K, Zosso D. Variational mode decomposition. IEEE Trans Signal Process, 2014, 62(3): 531-544

[30] Li HT, Jin K, Sun SL, Jia XY, Li YW (2022) Metro passenger flow forecasting though multi-source time-series fusion: An ensemble deep learning approach. Appl Soft Comput 120:108644

[31] Li J, Zhang ZC, Meng FX, Zhu W (2022) Short-term traffic flow prediction via improved mode decomposition and self-attention mechanism based deep learning approach. IEEE Sens J 22(14):14356–14365

[32] Yang H, Cheng YX, Li GH (2022) A new traffic flow prediction model based on cosine similarity variational mode decomposition, extreme learning machine and iterative error compensation strategy. Eng Appl Artif Intell 115:105234

[33] Zhang YY, Zhu CF, Wang QR (2020) LightGBM-based model for metro passenger volume forecasting. IET Intell Transp Syst 14(13):1815–1823

[34] Fu WL, Wang K, Li CS, Tan JW (2019) Multi-step short-term wind speed forecasting approach based on multi-scale dominant ingredient chaotic analysis, improved hybrid GWO-SCA optimization and ELM. Energy Convers Manag 187:356–377

[35] Huang YS, Gao YL, Gan Y, Ye M (2021) A new financial data forecasting model using genetic algorithm and long short-term memory network. Neurocomputing 425:207–218

[36] Liu Q, Liu M, Zhou HL, Yan F (2022) A multi-model fusion based non-ferrous metal priceforecasting. Resour Policy 77:102714

[37] Li GH, Zheng CF, Yang H (2022) Carbon price combination prediction model based on improved variational mode decomposition. Energy Rep 8:1644–1664

[38] Yang K, Wang BF, Qiu X, Li JH, Wang YZ, Liu YL (2022) Multi-step short-term wind speed prediction models based on adaptive robust decomposition coupled with deep gated recurrent unit. Energies 15(12):4221

[39] Abualigah L, Diabat A, Mirjalili S, Abd Elaziz M, Gandomi AH (2021) The arithmetic optimization algorithm. Comput Methods Appl Mech Eng 376:113609

[40] Goh HH, He RH, Zhang DD, Liu H, Dai W, Lim CS, Kurniawan TA, Teo KTK, Goh KC (2022) A multimodal approach to chaotic renewable energy prediction using meteorological and historical information. Appl Soft Comput 118:108487

[41] Sun W, Duan M. Analysis and forecasting of the carbon price in China’s regional carbon markets based on fast ensemble empirical mode decomposition, phase space reconstruction, and an improved extreme learning machine. Energies, 2019, 12(2): 277

[42] Li HT, Jin F, Sun SL, Li YW (2021) A new secondary decomposition ensemble learning approach for carbon price forecasting. Knowl Based Syst 214:106686

[43] Azami H, Escudero J. Amplitude- and fluctuation-based dispersion entropy. Entropy, 2018, 20(3): 210

[44] Chen C, Liu H. Dynamic ensemble wind speed prediction model based on hybrid deep reinforcement learning. Adv Eng Inform, 2021, 48

[45] Dong YC, Zhang HL, Wang C, Zhou XJ (2021) Wind power forecasting based on stacking ensemble model, decomposition and intelligent optimization algorithm. Neurocomputing 462:169–184

[46] Liu H, Deng DH (2022) An enhanced hybrid ensemble deep learning approach for forecasting daily PM2.5. J Cent South Univ 29(6):2074–2083

[47] Tian ZD. Modes decomposition forecasting approach for ultra-short-term wind speed. Appl Soft Comput, 2021, 105

[48] Song JJ, Wang JZ, Lu HY (2018) A novel combined model based on advanced optimization algorithm for short-term wind speed forecasting. Appl Energy 215:643–658

[49] Cheng ZS, Wang JY (2020) A new combined model based on multi-objective salp swarm optimization for wind speed forecasting. Appl Soft Comput 92:106294

[50] Chen J, Liu H, Chen C, Duan Z. Wind speed forecasting using multi-scale feature adaptive extraction ensemble model with error regression correction. Expert Syst Appl, 2022, 207

[51] Liu H, Yu CQ, Wu HP, Duan Z, Yan GX (2020) A new hybrid ensemble deep reinforcement learning model for wind speed short term forecasting. Energy 202:117794

[52] Jin ZZ, He DQ, Lao ZP, Wei ZX, Yin XH, Yang WF (2023) Early intelligent fault diagnosis of rotating machinery based on IWOA-VMD and DMKELM. Nonlinear Dyn 111(6):5287–5306

[53] Liu Y, Qin Z, Liao XF, Wu JH (2020) Cryptanalysis and enhancement of an image encryption scheme based on a 1-D coupled Sine map. Nonlinear Dyn 100(3):2917–2931

[54] He DQ, Liu CY, Jin ZZ, Ma R, Chen YJ, Shan S (2022) Fault diagnosis of flywheel bearing based on parameter optimization variational mode decomposition energy entropy and deep learning. Energy 239:122108

[55] Liang T, Lu H, Sun HX (2021) Application of parameter optimized variational mode decomposition method in fault feature extraction of rolling bearing. Entropy 23(5):520

[56] Hochreiter S, Schmidhuber J. Long short-term memory. Neural Comput, 1997, 9(8): 1735-1780

[57] Sun W, Tan B, Wang QQ. Multi-step wind speed forecasting based on secondary decomposition algorithm and optimized back propagation neural network. Appl Soft Comput, 2021, 113

[58] Wiering MA, Otterlo MV. Reinforcement learning. Adapt Learn Optim, 2012, 12(3): 729

[59] Szepesvári C. Algorithms for reinforcement learning. Synth Lect Artif Intell Mach Learn, 2010, 4(1): 1-103.

[60] Li XW, Huang ZX, Liu SH, Wu JX, Zhang YX (2023) Short-term subway passenger flow prediction based on time series adaptive decomposition and multi-model combination (IVMD-SE-MSSA). Sustainability 15(10):7949

[61] Yang XS (2010) A new metaheuristic bat-inspired algorithm. In: Proceedings of the international workshop on nature inspired cooperative strategies for optimization, pp 65–74

[62] Deb K, Pratap A, Agarwal S, Meyarivan T. A fast and elitist multiobjective genetic algorithm: NSGA-II. IEEE Trans Evol Comput, 2002, 6(2): 182-197

Funding

National Natural Science Foundation of China(No. 52072081)

Major Project of Science and Technology of Guangxi Province of China(No. Guike AA20302010)

Interdisciplinary Scientific Research Foundation of Guangxi University(No. 2022JCA003)

Innovation Project of Guangxi Graduate Education(No. YCSW2023086)