Prediction of the Inventory of New Energy Vehicle Charging Piles in China Based on Support Vector Machine

Jun Li; Yumei Wang

doi:10.63313/EPP.9001

Authors

Jun Li Anhui University of Finance and Economics, Bengbu 233030, China Author
Yumei Wang Anhui University of Finance and Economics, Bengbu 233030, China Author

DOI:

https://doi.org/10.63313/EPP.9001

Keywords:

New Energy Vehicles, Charging Pile Inventory, Support Vector Machine, Predic-tion Model

Abstract

With the rapid development of China's new energy vehicle (NEV) industry, the charging pile, as a key supporting facility, plays a crucial role. Predicting the inventory of charging piles is of great significance for guiding the healthy de-velopment of the industry. This paper employs the Support Vector Machine (SVM) as the prediction model. Firstly, data on the inventory of NEV charging piles in China over the past six years were collected. Combined with factors such as the production and sales volume of NEVs, relevant policies, and public pa-tents, a prediction model was constructed. By selecting appropriate kernel func-tions and adjusting parameters, the SVM model demonstrated good predictive performance on the training data. Cross-validation and parameter optimization were conducted to ensure the stability and reliability of the prediction results. The prediction indicates that the inventory of NEV charging piles in China is expected to continue growing in the coming years.

References

[1] Bae, S. and Kwasinski, A. (2012) Electric Vehicle Charging Infrastructure Planning: A Re-view of Modeling Approaches. Renewable and Sustainable Energy Reviews, 16, 5143-5152.

https://doi.org/10.1016/j.rser.2012.05.022

[2] Gnann, M. and Plötz, P. (2019) Forecasting the Demand of Electric Vehicles for Private and Shared Mobility. Transportation Research Part D: Transport and Environment, 71, 156-172.

https://doi.org/10.1016/j.trd.2018.12.011

[3] Chen, H., Luh, P.B. and Zhao, C. (2004) Support Vector Regression for Electric Load Fore-casting. IEEE Transactions on Power Systems, 19, 1825-1834.

https://doi.org/10.1109/TPWRS.2004.835637

[4] Al-Sumaiti, A.K., Ahmed, M.H. and Salama, M. (2020) A Review of Machine Learning Appli-cations in Energy Systems. IEEE Access, 8, 212518-212541.

https://doi.org/10.1109/ACCESS.2020.3040394

[5] Liu, J., Chen, H. and Zhang, X. (2021) Charging Infrastructure Demand Modeling for Plug-in Electric Vehicles: A Review of Approaches and Methods. Energy, 223, 120044.

https://doi.org/10.1016/j.energy.2021.120044

[6] Li, Y., Zhang, L. and Li, W. (2020) Optimal Planning of Electric Vehicle Charging Stations Considering Traffic Flow and Power Grid Constraints. Applied Energy, 278, 115654.

https://doi.org/10.1016/j.apenergy.2020.115654

[7] Wang, Y., Infield, D. and Wang, Y. (2021) Short-Term Electric Vehicle Charging Demand Prediction Using Deep Learning Methods. Energy, 214, 118874.

https://doi.org/10.1016/j.energy.2020.118874

[8] Zhang, L., Brown, T. and Samuelsen, G.S. (2018) Evaluation of Charging Infrastructure Re-quirements for Plug-in Electric Vehicles under Different Market Penetration Scenari-os. Transportation Research Part D: Transport and Environment, 65, 591-611.

https://doi.org/10.1016/j.trd.2017.10.006

[9] Smets, K., Verdonk, B. and Jordaan, E.M. (2019) Support Vector Machines for Regression: A Comparison of Optimization Techniques. Neural Computing and Applications, 31, 2603-2615.

https://doi.org/10.1007/s00521-017-3214-2

[10] Dong, M., Liu, C. and Lin, Z. (2022) Policy Incentives and the Adoption of Electric Vehicles: A Global Review. Renewable and Sustainable Energy Reviews, 156, 111939.

https://doi.org/10.1016/j.rser.2021.111939