AI and machine learning applications in energy efficiency

Abstract

The integration of Artificial Intelligence (AI) and Machine Learning (ML) into energy management systems has demonstrated remarkable potential for improving energy efficiency across various sectors. This study explores the application of ML models, including Linear Regression, Random Forest Regression, and Artificial Neural Networks, to predict energy consumption and optimize scheduling for minimizing peak loads. Historical energy usage data were analyzed to train and validate the models, with performance evaluated using metrics such as Mean Absolute Error (MAE) and Root Mean Square Error (RMSE). The findings reveal that ANN outperformed other models, achieving an MAE of 5.3 and RMSE of 6.8, closely aligning with actual energy usage patterns. Furthermore, optimization techniques resulted in significant energy savings, reducing consumption from 1200 kWh to 950 kWh. These results highlight the transformative role of AI and ML in achieving sustainability goals by enhancing energy efficiency, reducing costs, and mitigating environmental impact.