A specification of neural network applications in the load forecasting problem

Abstract

This paper investigates the effectiveness of the Artificial Neural Network (ANN) approach to short term load forecasting in electrical power systems. Using examples, the learning process and capabilities of a neural network in the prediction of peak load of the day are demonstrated. Different data normalizing approaches and input patterns are employed to exploit the correlation between historical load and temperatures and expected load patterns. A number of AN"s are included with emphasis given to their practical implementation for electrical power system control and planning purposes. The networks have been trained on actual power utility load data using a backpropagation Algorithm. The prospects for applying a combined solution using artificial neural networks and expert systems, called the expert network are also discussed. Consideration is given to expert networks as a more complete solution to the forecasting problem which neither system alone can provide.

Domestic Heat Demand Prediction Using Neural Networks

“…Although electricity demand prediction is studied quite extensively [2,1], individual heat demand prediction is an unexplored field. Most demand prediction schemes try to predict the demand for a large area, for example a complete neighborhood.…”

Comparison neural networks models for short term forecasting of natural gas consumption in Istanbul

“…Neural network techniques have been recently suggested for short-term, midterm or long term load forecasting by a large number of researchers [6][7][8][9][10][11][12][13]. As the models presented in the literature are either intended for forecasting the whole daily load curve at once, or forecasting the load weekly, this division is used here in testing different models.…”

Modeling and forecasting electric daily peak loads using abductive networks

“…Ref. [6] gives results on 12 dedicated neural network models that predict the peak load on the working day Thursday using various combinations of input variables and data normalization schemes. The MAPE error obtained varied from 1.96% to 4.80%, with 10 out of the 12 models giving MAPE ≥ 2.37%.…”

“…Feed-forward neural networks trained with error back-propagation have been widely used for modeling and forecasting the daily peak load, e.g. [6][7][8][9][10][11][12][13]. However, neural networks suffer from a number of limitations, including difficulty in determining optimum network topology and training parameters [14].…”