Regularization of neural networks for improved load forecasting in power system

Presents the regularization procedure for the neural network reduction to obtain the best results of load forecasting in the power system. The OBD pruning method will be applied in the solution. The numerical experiments have been concentrated on the prognosis of the load in the power system. Two kinds of experiments are described: 24-hour forecast and the forecast of the daily mean of the load. It will be shown that application of the regularization of the neural network employed for prediction will result in significant improvement of the forecasting accuracy.