The European Union's long-term objectives have made electricity demand forecasting a top priority. Lack of adequate methods for estimating future electricity needs, which leads to either under- or over-investment in energy infrastructure. These problems can be solved by employing predictive analysis and time series forecasting techniques. In this study, we will use the Seasonal ARIMA Integrated Moving Average (SARIMA) to make short-term predictions about power consumption and then compare those predictions to those made using MLP stands for "Multi-Layer Perceptron," which is the name of the algorithm used. The electricity consumption in London is evaluated using a half-hourly dataset gathered from UK power networks between November 2011 and February 2014, which includes information on weather and holidays in the city. Through the use of forecasting plots based on the highest, lowest, and middle points of consumption, the MAE, MAPE, MSE, and RMSE of SARIMA and MLP were calculated (RMSE). The prediction graphs are shown on the user interface, and MLP fared better than SARIMA in the tests.