Fog-based AI image analysis for load disaggregation using Random Forest and XGBoost

The use of artificial intelligence to energy systems is growing, and image-based analysis is showing promise as a tool to improve optimization and monitoring. This paper presents a new load disaggregation architecture that improves responsiveness and efficiency by combining fog computing with artificial intelligence-driven image recognition approaches. To categorize consumption profiles the system examines visual depictions of energy data. Two different kinds of consumption profiles were identified and differentiated using Random Forest and XGBoost classifiers, allowing precise energy usage disaggregation in complex contexts.