A Comprehensive Survey on AI-Based Smart EV Charging Station with Load Prediction

This survey presented a comprehensive view of recent research works on AI-based smart EV charging stations with load prediction. The reviewed approaches were categorised based on machine learning, deep learning and reinforcement learning methods. The machine learning category covered XGBoost [15], Random Forest [16] and LightGBM [17] methods for improving the forecasting efficiency. The deep learning category included LSTM [18], CNN-LSTM [19] and Transformer based [20] architectures for advancing temporal modelling. The reinforcement learning category included multi-objective DRL [21], multi-agent graph RL [22] and ANN-based energy management [23] for optimising the charging schedule. The methodologies, advantages and limitations of the reviewed works were compared. The existing techniques overcome several issues such as load forecasting error, demand response and cost efficiency. However, downsides such as cold-start performance, computational complexity, interpretability and privacy still persist. Future research is expected to focus on explainable AI, federated learning and physics-informed hybrid models for the realisation of robust, scalable and trustworthy AI-based smart EV charging infrastructure.

[1] International Energy Agency, "Global EV Outlook 2024: Trends in electric vehicle adoption," IEA, Paris, France, 2024. [2] T. Mastoi, S. Zhuang, H. M. Munir, M. Haris, M. Hassan, M. Usman, S. S. H. Bukhari and J. Ro, "An in-depth analysis of electric vehicle charging station infrastructure, policy implications, and future trends," Energy Reports, vol. 9, pp. 1777–1806, 2023. [3] A. R. Singh, R. S. Kumar, K. R. Madhavi, F. Alsaif, M. Bajaj and I. Zaitsev, "Optimizing demand response and load balancing in smart EV charging networks using an AI integrated blockchain framework," Scientific Reports, vol. 14, pp. 1–20, 2024. [4] A. F. Güven, N. Ateş, S. Alotaibi, T. Alzahrani, A. M. Amsal and S. K. Elsayed, "Sustainable hybrid systems for electric vehicle charging infrastructures in regional applications," Scientific Reports, vol. 15, 2025. [5] S. Mahmud, M. T. Iqbal and S. Mekhilef, "A comprehensive review of electric vehicle charging stations with solar photovoltaic integration," Renewable and Sustainable Energy Reviews, vol. 180, p. 113311, 2023. [6] M. Shahriar, A. R. Al-Ali, A. H. Osman, S. Dhou and M. Nijim, "Machine learning approaches for EV charging behaviour: a review," IEEE Access, vol. 8, pp. 168980–168993, 2020. [7] S. Mohammadi, R. Bayindir and S. Padmanaban, "A survey on deep learning techniques for energy management in electric vehicle charging stations," Energy Reports, vol. 11, pp. 2200–2218, 2025. [8] K. R. Ullah, A. Ahmad and R. M. Khan, "Charging stations demand forecasting using LSTM based hybrid Transformer model," Scientific Reports, vol. 15, 2025. [9] M. Abdullah-Al-Nahid, T. Aziz Khan, A. Ahmed and M. Z. R. Khan, "Reinforcement learning for electric vehicle charging management: theory and applications," Energies, vol. 18, no. 19, p. 5225, 2025. [10]                 A. Sharma, S. Rani and M. Shabaz, "Artificial intelligence-augmented smart grid architecture for cyber intrusion detection and mitigation in electric vehicle charging infrastructure," Scientific Reports, vol. 15, 2025.