Comparison of Feasibility between Machine Learning Algorithms In Terms of Predicting Energy Consumption of Smart Grid

Electricity is one of the most important things. There are so many sectors in where we need the supply of electricity continuously. Smart Grid is an automated system which can automatically control the supply of the electricity according to demand. Artificial Intelligence and Machine Learning is essential for Smart Grid’s mechanism. So in this paper we have compared the feasibility of three Machine Learning Algorithms. We have investigated about Support Vector Machine (SVM), Random Forest, Linear Regression. The feasibility was measured in terms of Root Mean Square(RMSE) and Mean Absolute (MAE). We have applied these three Machine Learning methods on a data-set of electricity usages in 2-storied building in USA. By comparing, we can decide which Machine Learning method will be most feasible for Smart Grid.