Forecasting the electricity demand using machine learning algorithms

No Thumbnail Available
Shaik, Safdhar Ali
Issue Date
MSc in Data Analytics
Dublin Business School
Items in eSource are protected by copyright. Previously published items are made available in accordance with the copyright policy of the publisher/copyright holder.
Forecasting the electricity demand has been the key task to reach the long-term goals of European Union. Inadequacy of accurate measurement of electricity demand which further causes the under generation or over generation of electricity and may also results in huge investments on energy resources. Predictive analysis and time series forecasting methods are used to overcome these difficulties. The aim of this paper is to short-term forecast the electricity demand using the seasonal auto-regressive Integrated moving average (SARIMA) and compare the results with the Multi-layer perceptron (MLP) model. The half-hourly dataset of London city is used for analyzing the electricity demand which is collected from UK power networks along with weather and holidays in the same city from November 2011 to February 2014. The forecasting plots has made based on the maximum, minimum and average consumption and then compare the mean absolute error (MAE), mean absolute percentage error (MAPE), mean square error (MSE) and root mean square error (RMSE) of SARIMA with the MLP model. During the evaluation MLP outperformed the SARIMA model and the prediction graphs are displayed using User Interface.