RADIAL BASIS FUNCTION NEURAL NETWORKS-BASED SHORT TERM ELECTRIC POWER LOAD FORECASTING FOR SUPER HIGH VOLTAGE POWER GRID

Load forecasting plays an essential role both in developed and developing countries for policymakers and related organizations. It helps an electrical utility to make important decisions including decisions on purchasing and generating electrical power, load switching, and infrastructure development. In recent years Artificial Neural Networks (ANNs) have been applied for short-term power load forecasting (STPLF). This work presents a study of STPLF for the Iraqi national grid by means of Radial Basis Function NN(RBFNN) and Multi-Layer Perceptron NN (MLPNN) model. Inputs to the ANN are past loads and the output of the ANN is the load forecast for given days. Historical load data obtained from the Control and Operation Office at the Iraqi ministry of electricity has been split into two main parts, where 50% of the data are used for the training and the other 50% has been devoted to test the trained network. Simulations have been accomplished in MATLAB environment, where the data have been preprocessed and rearranged. Lastly, the simulation results proved that the predicted load values are following closely the actual load.