Use of Artificial Neural Networks for Short-Term Electricity Load Forecasting of Kenya National Grid Power System

Abstract

This paper developed a supervised Artificial Neural Network-based model for Short-Term Electricity Load Forecasting, and evaluated the performance of the model by applying the actual load data of the Kenya National Grid power system to predict the load of one day in advance. Raw data was collected, cleaned and loaded onto the model. The model was trained under the WEKA environment and predicted the total load for Kenya National Grid power system. The test results showed that the hour-by-hour approach is more suitable and efficient for a day-ahead load forecasting. Forecast results demonstrated that the model performed remarkably well with increased number of iterations. The result suggests that incremental training approach of a neural network model should be implemented for online testing application to acquire a universal final view on its applicability. General Terms Load Forecasting.