Modeling of Electric Power Consumption: Evaluating the Efficacy of Linear, Polynomial, SVR, and Random Forest Methods

In this research, we are focusing on the analysis of electric power consumption, while using different machine-learning models. Different algorithms have been used, such as linear regression, polynomial regression, support vector regression (SVR), artificial neural network (multi-layer perceptron), and random forest, to identify the model for the most efficient electric power consumption prediction in different zones. Every model is graded with the following metric: Mean Squared Error (MSE), Root Mean Squared Error (RMSE), Mean Absolute Error (MAE), and R-squared Score. The results indicate that random forest achieved the highest performance with an R-squared value of 97%, followed closely by SVR and multi-layer perceptron both at 90% and polynomial regression at 86%. In contrast, linear regression yielded a lower R squared value of 75%, suggesting it is not well-suited for addressing this issue.