@Article{jai.2025.066067,
AUTHOR = {Mohammed Moawad Alenazi},
TITLE = {Machine Learning-Optimized Energy Management for Resilient Residential Microgrids with Dynamic Electric Vehicle Integration},
JOURNAL = {Journal on Artificial Intelligence},
VOLUME = {7},
YEAR = {2025},
NUMBER = {1},
PAGES = {143--176},
URL = {http://www.techscience.com/jai/v7n1/62800},
ISSN = {2579-003X},
ABSTRACT = {This paper presents a novel machine learning (ML) enhanced energy management framework for residential microgrids. It dynamically integrates solar photovoltaics (PV), wind turbines, lithium-ion battery energy storage systems (BESS), and bidirectional electric vehicle (EV) charging. The proposed architecture addresses the limitations of traditional rule-based controls by incorporating ConvLSTM for real-time forecasting, a Twin Delayed Deep Deterministic Policy Gradient (TD3) reinforcement learning agent for optimal BESS scheduling, and federated learning for EV charging prediction—ensuring both privacy and efficiency. Simulated in a high-fidelity MATLAB/Simulink environment, the system achieves 98.7% solar/wind forecast accuracy and 98.2% Maximum Power Point Tracking (MPPT) tracking efficiency, while reducing torque oscillations by 41% and peak demand by 22%. Compared to baseline methods, the solution improves voltage and frequency stability (maintaining 400 V ±2%, 50 Hz ±0.015 Hz) and achieves a 70% reduction in battery State of Charge (SOC) management error. The EV scheduler, informed by data from over 500 households, reduces charging costs by 31% with rapid failover to critical loads during outages. The architecture is validated using ISO 8528-8 transient tests, demonstrating 99.98% uptime. These results confirm the feasibility of transitioning microgrids from reactive systems to adaptive, cognitive infrastructures capable of self-optimization under highly variable renewable generation and EV behaviors.},
DOI = {10.32604/jai.2025.066067}
}