Predicting Parking Spaces Using CEEMDAN and GRU

doi:10.1007/s12204-023-2672-1

Abstract

Abstract: Accurate prediction of parking spaces plays a crucial role in maximizing the efficiency of parking resources and optimizing traffic conditions. However, the majority of earlier research has used models based on past parking data or the plethora of variables that influence parking prediction, which not only makes the data more complicated and costs more time to run but can also lead to poor model fits. To solve this problem, a hybrid parking prediction model combining complete ensemble empirical mode decomposition with adaptive noise (CEEMDAN) and gated recurrent unit (GRU) model is proposed to predict the number of parking spaces. In this model, CEEMDAN has the ability to gradually break down time series fluctuations or trends at various scales, producing a sequence of intrinsic mode functions (IMF) with various characteristic scales. Then, by keeping the majority of the original data’s content, removing superfluous information, and enhancing predicted response time, principal component analysis (PCA) decreases the dimensionality of the IMF series. Subsequently, the high-level abstract characteristics are entered into the GRU network, and the network is built, tested, and predicted based on the deep learning framework Keras. The validity of the presented model is verified by making use of real parking datasets from two three-dimensional parking lots. The test results reveal that the model outperforms the baseline model’s predictive accuracy, i.e., a lower testing error. The real parking time series are most closely modeled by the CEEMDAN-PCA-GRU model. As a result, the method is superior to existing models for parking prediction.

Key words: parking prediction, principal component analysis (PCA), deep learning, complete ensemble empirical mode decomposition with adaptive noise (CEEMDAN), gated recurrent unit (GRU), time series

摘要： 剩余停车位的精准预测对于优化停车资源利用率、改善交通状况起着至关重要的作用。然而，以往的研究大多基于停车本身的历史数据或影响停车预测的众多因素进行模型建模，增加了数据的复杂性和运行模型所花费的时间，导致模型与极值点的拟合度较差。针对这一问题，提出一种基于完全自适应噪声集合经验模态分解（CEEMDAN）和门循环单元（GRU）模型的混合预测模型来预测停车剩余位数。在该模型中，CEEMDAN作为序列平滑分解模块，可以逐步分解不同尺度的时间序列波动或趋势，生成一系列具有不同特征尺度的本征模态函数（IMF）。然后，通过保留原始数据的大部分信息，主成分分析（PCA）减少了分解的IMF序列维度，消除了冗余信息，提高了预测响应速度。之后，高级抽象特征输入GRU网络，基于深度学习框架Keras完成网络的搭建、测试、预测。利用立体停车场采集的真实停车数据集验证了所提模型的有效性。实验结果表明，本文提出的模型在预测准确性方面优于基准模型，即更低的测试误差。CEEMDAN-PCA-GRU模型最接近真实的停车时间序列。因此，该方法在停车位预测方面比其他模型更有效。

关键词: 停车预测，主成分分析（PCA），深度学习，完全自适应噪声集合经验模态分解（CEEMDAN），门循环单元（GRU），时间序列

CLC Number:

U491.7

MA Changxi, HUANG Xiaoting, MENG Wei. Predicting Parking Spaces Using CEEMDAN and GRU[J]. J Shanghai Jiaotong Univ Sci, 2025, 30(5): 962-975.

References

[1] AFRIN T, YODO N. A survey of road traffic congestion measures towards a sustainable and resilient transportation system [J]. Sustainability, 2020, 12(11): 4660.

[2] KOUMETIO TEKOUABOU S C, ABDELLAOUI ALAOUI E A, CHERIF W, et al. Improving parking availability prediction in smart cities with IoT and ensemble-based model [J]. Journal of King Saud University - Computer and Information Sciences, 2022, 34(3): 687-697.

[3] PARMAR J, DAS P, DAVE S M. Study on demand and characteristics of parking system in urban areas: A review [J]. Journal of Traffic and Transportation Engineering (English Edition), 2020, 7(1): 111-124.

[4] CAICEDO F. The use of space availability information in “PARC” systems to reduce search times in parking facilities [J]. Transportation Research Part C: Emerging Technologies, 2009, 17(1): 56-68.

[5] LIN T, RIVANO H, LE MOUËL F. A survey of smart parking solutions [J]. IEEE Transactions on Intelligent Transportation Systems, 2017, 18(12): 3229-3253.

[6] JI Y J, GAO L P, CHEN X S, et al. Strategies for multi-step-ahead available parking spaces forecasting based on wavelet transform [J]. Journal of Central South University, 2017, 24(6): 1503-1512.

[7] ZENG C, MA C X, WANG K, et al. Predicting vacant parking space availability: A DWT-Bi-LSTM model [J]. Physica A: Statistical Mechanics and Its Applications, 2022, 599: 127498.

[8] ABEDINIA O, LOTFI M, BAGHERI M, et al. Improved EMD-based complex prediction model for wind power forecasting [J]. IEEE Transactions on Sustainable Energy, 2020, 11(4): 2790-2802.

[9] LI G X, ZHONG X. Parking demand forecasting based on improved complete ensemble empirical mode decomposition and GRU model [J]. Engineering Applications of Artificial Intelligence, 2023, 119: 105717.

[10] HUANG Y A, YU J H, DAI X H, et al. Air-quality prediction based on the EMD–IPSO–LSTM combination model [J]. Sustainability, 2022, 14(9): 4889.

[11] ZHANG Y A, YAN B B, AASMA M. A novel deep learning framework: Prediction and analysis of financial time series using CEEMD and LSTM [J]. Expert Systems With Applications, 2020, 159: 113609.

[12] 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.

[13] TANG J J, CHEN X Q, HU Z, et al. Traffic flow prediction based on combination of support vector machine and data denoising schemes [J]. Physica A: Statistical Mechanics and Its Applications, 2019, 534: 120642.

[14] CHEN X Q, LU J Q, ZHAO J S, et al. Traffic flow prediction at varied time scales via ensemble empirical mode decomposition and artificial neural network [J]. Sustainability, 2020, 12(9): 3678.

[15] CAO J, LI Z, LI J. Financial time series forecasting model based on CEEMDAN and LSTM [J]. Physica A: Statistical Mechanics and Its Applications, 2019, 519: 127-139.

[16] ZHOU F T, HUANG Z H, ZHANG C H. Carbon price forecasting based on CEEMDAN and LSTM [J]. Applied Energy, 2022, 311: 118601.

[17] WANG J, CAO J X, YUAN S, et al. Short-term forecasting of natural gas prices by using a novel hybrid method based on a combination of the CEEMDAN-SE-and the PSO-ALS-optimized GRU network [J]. Energy, 2021, 233: 121082.

[18] KUMAR S V, VANAJAKSHI L. Short-term traffic flow prediction using seasonal ARIMA model with limited input data [J]. European Transport Research Review, 2015, 7(3): 1-9.

[19] 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.

[20] RAJABIOUN T, IOANNOU P A. On-street and off-street parking availability prediction using multivariate spatiotemporal models [J]. IEEE Transactions on Intelligent Transportation Systems, 2015, 16(5): 2913-2924.

[21] XIAO J, LOU Y Y, FRISBY J. How likely am I to find parking? A practical model-based framework for predicting parking availability [J]. Transportation Research Part B: Methodological, 2018, 112: 19-39.

[22] AWAN F M, SALEEM Y, MINERVA R, et al. A comparative analysis of machine/deep learning models for parking space availability prediction [J]. Sensors, 2020, 20(1): 322.

[23] MEI Z Y, ZHANG W, ZHANG L H, et al. Real-time multistep prediction of public parking spaces based on Fourier transform–least squares support vector regression [J]. Journal of Intelligent Transportation Systems, 2020, 24(1): 68-80.

[24] INAM S, MAHMOOD A, KHATOON S, et al. Multisource data integration and comparative analysis of machine learning models for on-street parking prediction [J]. Sustainability, 2022, 14(12): 7317.

[25] FAN J K, HU Q, TANG Z Z. Predicting vacant parking space availability: An SVR method with fruit fly optimisation [J]. IET Intelligent Transport Systems, 2018, 12(10): 1414-1420.

[26] YE X F, WANG J F, WANG T, et al. Short-term prediction of available parking space based on machine learning approaches [J]. IEEE Access, 2020, 8: 174530-174541.

[27] JELEN G, PODOBNIK V, BABIC J. Contextual prediction of parking spot availability: A step towards sustainable parking [J]. Journal of Cleaner Production, 2021, 312: 127684.

[28] WU Y K, TAN H C. Short-term traffic flow forecasting with spatial-temporal correlation in a hybrid deep learning framework [DB/OL]. (2016-12-03). https://arxiv.org/abs/1612.01022

[29] PICCIALLI F, GIAMPAOLO F, PREZIOSO E, et al. Predictive analytics for smart parking: A deep learning approach in forecasting of IoT data [J]. ACM Transactions on Internet Technology, 2021, 21(3): 1-21.

[30] ZENG C, MA C X, WANG K, et al. Parking occupancy prediction method based on multi factors and stacked GRU-LSTM [J]. IEEE Access, 2022, 10: 47361-47370.

[31] ZHANG W J, LIU H, LIU Y C, et al. Semi-supervised city-wide parking availability prediction via hierarchical recurrent graph neural network [J]. IEEE Transactions on Knowledge and Data Engineering, 2022, 34(8): 3984-3996.

[32] YANG S G, MA W, PI X D, et al. A deep learning approach to real-time parking occupancy prediction in transportation networks incorporating multiple spatio-temporal data sources [J]. Transportation Research Part C: Emerging Technologies, 2019, 107: 248-265.

[33] FAN J K, HU Q, XU Y Y, et al. Predicting vacant parking space availability: A long short-term memory approach [J]. IEEE Intelligent Transportation Systems Magazine, 2022, 14(2): 129-143.

[34] FENG Y J, XU Y Y, HU Q, et al. Predicting vacant parking space availability zone-wisely: A hybrid deep learning approach [J]. Complex & Intelligent Systems, 2022, 8(5): 4145-4161.

[35] GAO L P, FAN W L, HU Z Y, et al. Prediction of vacant parking spaces in multiple parking lots: A DWT-ConvGRU-BRC model [J]. Applied Sciences, 2023, 13(6): 3791.

[36] GROTH D, HARTMANN S, KLIE S, et al. Principal components analysis [M]// Computational toxicology. Totowa: Humana Press, 2013: 527-547.

[37] LIU Z G, LI W J, FENG J X, et al. Research on satellite network traffic prediction based on improved GRU neural network [J]. Sensors, 2022, 22(22): 8678.