Authors:
K. Manish Stephen, V. Krishnakumar, B. Ranesh Karthik, R. Angeline, Dilli Kasi Rao Kotha

Addresses:
Department of Computer Science and Engineering in Artificial Intelligence and Machine Learning, SRM Institute of Science and Technology, Ramapuram, Chennai, Tamil Nadu, India. Faculty of Engineering, Environment and Computing, Coventry University, Coventry, England, United Kingdom.

Abstract:

Intelligent, scalable, and self-sufficient energy management is needed when localised energy systems incorporate solar and wind energy. Most classic microgrid control methods rely on reactive, static optimisation and human intervention, rendering them vulnerable to changes in renewable energy generation, load variations, and information uncertainty. Some current solutions rely on hardware- or cloud-based architectures, which are inflexible, increase latency, and lack reliability due to communication constraints. This paper proposes an intelligent energy management system, AetherGrid. This software-based technology operates autonomously in high-dynamic, uncertain renewable energy conditions. AetherGrid values post-data-capture intelligent processing, including feature extraction, prediction, and optimisation, utilising hybrid machine learning models. This system continuously analyses energy data to forecast demand, renewable energy output, and system state, enabling intelligent decision-making. This system adapts to new consumption trends and environmental conditions without affecting performance or stability, regardless of whether it uses hardware-dependent regulatory mechanisms or cloud services. The layers that handle and analyse information in AetherGrid become intelligent, making it a future smart energy solution. Home energy management systems, campus-scale systems, and other applications benefit from the technology. This research shows that energy management systems can become intelligent through software-based artificial intelligence.

Keywords: Renewable Energy; Energy Management; Artificial Intelligence; Machine learning; Demand Prediction; Smart Grids; Edge Computing; Time-Series Prediction; Feature Extraction.

Received on: 23/01/2025, Revised on: 28/03/2025, Accepted on: 27/06/2025, Published on: 03/03/2026

DOI: 10.69888/FTSESS.2026.000689

FMDB Transactions on Sustainable Environmental Sciences, 2026 Vol. 3 No. 1, Pages: 1-11

• Views : 49
• Downloads : 11

Download PDF