Edge-Intelligent Electric Vehicle Charging Coordination for Grid Load Balancing and Renewable Integration
Yazarlar (2)
Zafer Dogan Tokat Gaziosmanpaşa Üniversitesi
| Makale Türü | Özgün Makale (Uluslararası alan indekslerindeki dergilerde yayınlanan tam makale) | ||
| Dergi Adı | Cluster Computing - The Journal of Networks, Software Tools and Applications | ||
| Makale Dili | – | Basım Tarihi | 01-2025 |
| Makale Linki | https://papers.ssrn.com/sol3/papers.cfm?abstract_id=5251124 | ||
| UAK Araştırma Alanları |
Akıllı Elektrik Şebekeleri
|
||
| Özet |
| The rising adoption of electric vehicles (EVs) presents significant challenges for grid balancing and renewable integration. Traditional cloud-based EV charging systems suffer from latency and limited adaptability. This study introduces an edge-intelligent framework leveraging IoT-enabled sensing and on-site AI inference to coordinate real-time EV charging and optimize grid performance. Lightweight models (CNN+ LSTM, XGBoost, Random Forest) are deployed on Jetson Orin Nano and Raspberry Pi 5, enabling behavior-aware scheduling and dynamic pricing at the edge. Trained on a US Department of Energy dataset, the system achieves a 27.6% improvement in station utilization, 24.5% reduction in peak load, and 29.8% cost savings while maintaining high prediction accuracy (R2> 0.92). Communication overhead is reduced by 80%, supporting faster, decentralized decision-making. The proposed framework offers a scalable, practical solution for intelligent EV charging in sustainable smart grids. |
| Anahtar Kelimeler |
