Machine learning algorithms to forecast an electrical appliances consumption in dwellings

Predicting energy consumption in residential structures is a crucial topic of study due to increased demand for energy-efficient solutions. The electric utility sector relies on load estimates to balance power supply and ensure customers receive quality electricity. Understanding and predicting electricity consumption can help consumers organize and control their expenses. This is especially important for off-grid systems where power cuts disrupt scheduled activities and supplementing with other energy sources is not feasible or cost-effective. This article examines an individual dwelling's power use and employs statistical and machine learning (ML) methods to develop supervised approaches in order to forecast the span worth using. The developed load prediction models forecast electric load demand for a day ahead. Current research work looks into various ML strategies namely, linear regression, random forest, decision tree, multilayer perceptron, support vector regression and XGBoost regressor to compute the amount used of electric energy in domestic construction. This article employs MAE and RMSE metrics to monitor the model’s performance.