using machine learning algorithms to predict the electric daily peak demand

This paper aims to predict the maximum electric energy demand that is helpful to ease the appropriate arrangement of effective situations (such as the scheduling of eliminating some consumers). in this study, there are 24 features, the backward, forward, and stepwise regressions are applied to find the significant features for a reduced model. the two algorithms (neural network and regression) are applied on the training set (60 percent) of the real data thentested with the test set data (40 percent). the results show that both proposed models can be useful but the regression model performs better than the neural network model to predict the next day’s maximum demand of the electric energy.