Smart Monitoring of Power Networks: Augmented Reality and Edge AI for Grid Resilience
Yazarlar (1)
Dr. Öğr. Üyesi Abdulkadir GÖZÜOĞLU Tokat Gaziosmanpaşa Üniversitesi, Türkiye
| Bildiri Türü | Tebliğ/Bildiri | Bildiri Dili | İngilizce |
| Bildiri Alt Türü | Tam Metin Olarak Yayınlanan Tebliğ (Uluslararası Kongre/Sempozyum) | ||
| Bildiri Niteliği | Alanında Hakemli Uluslararası Kongre/Sempozyum | ||
| Kongre Adı | 6th International Conference on Scientific and Academic Research | ||
| Kongre Tarihi | 19-03-2025 / 20-03-2025 | ||
| Basıldığı Ülke | Türkiye | Basıldığı Şehir | Konya |
| Bildiri Linki | https://drive.google.com/file/d/1GzUzFvR2nVvHhlqt2DuLW8pRFCQujTcZ/view | ||
| UAK Araştırma Alanları |
Akıllı Elektrik Şebekeleri
Yapay Zeka
Elektrik Enerjisi ve Güç Sistemleri
|
||
| Özet |
| The increasing complexity of modern power networks, driven by the rapid integration of renewable energy sources and decentralized energy systems, has introduced significant challenges in grid resilience, fault detection, and operational efficiency. Traditional SCADA-based monitoring approaches suffer from latency issues and centralized processing bottlenecks, leading to delayed fault detection and prolonged downtime. This paper presents a Smart Monitoring System that integrates Edge AI and Augmented Reality (AR) to enable real-time fault detection, predictive maintenance, and interactive visualization for power grid operators. The proposed system leverages Edge AI-driven predictive analytics to identify potential faults before they escalate into significant failures, enhancing decisionmaking capabilities through decentralized AI inference on edge devices. Furthermore, AR-based interfaces enable field technicians to visualize real-time grid data, remotely inspect high-voltage equipment, and receive AI-generated fault diagnosis overlaid onto live infrastructure. Our approach outperforms traditional SCADA systems, as Edge AI reduces fault detection latency by 56.7%. AR visualization improves operator response times by 37.8%, leading to a Grid Resilience Index (GRI) increase from 96.1% to 99.1%. Unlike conventional cloud-dependent models, our framework processes critical power grid data locally on Raspberry Pi 5-based Edge AI units, ensuring faster response times and enhanced fault mitigation. The results of our scenario-based analysis demonstrate that the proposed system achieves 95.4% fault detection accuracy, a significant … |
| Anahtar Kelimeler |
