Machine Learning Technique for the Prediction of Short-term Load Demand: A Case Study
Publication Date: 22/02/2023
Author(s): M.B. Jibril, Aliyu Sani, Usman Lawan Maska.
Volume/Issue: Volume 5 , Issue 1 (2023)
Abstract:
The purpose of this paper is to present a machine-learning approach for forecasting short-term load demand in Kano. Artificial Neural Network (ANN) and Support Vector Machine (SVM) are applied to develop the model. Three independent variables are selected as inputs, and one output is used to discover the level of relationship among the variables that are independent. This approach can ascertain a more precise prediction of the short-term load demand compared to expensive and rigorous experimental techniques. The correlation coefficient (R), coefficient of determination (R2), Mean Square Error (MSE), and Root Mean Square Error (RMSE) were used as indicators to evaluate the prediction accuracy of the selected algorithms. ANN gives a close accurate output as follows: R=0.97539, R2=0.951385, MSE=0.003674 and RMSE=0.060369.
Keywords:
Load demand forecasting, Back Propagation Neural Network (BPNN), Support Vector Machine (SVM).
