Load forecasting is very essential to the operation of electricity companies. It enhances the energy-efficient and reliable operation of a power system. This paper presents a study of short-term load forecasting using Artificial Neural Networks (ANNs) and applied it to the Nigeria Electric power system. This gives load forecasts one hour in advance. Historical load data obtained from the Power Holding Company of Nigeria (PHCN, formerly NEPA) for the month of August 2003 were used. The main stages are the pre-processing of the data sets, network training, and forecasting. The inputs used for the neural network are the previous hour load, previous day load, previous week load, day of the week, and hour of the day. The neural network used has 3 layers: an input, a hidden, and an output layer. The input layer has 5 neurons, the number of hidden layer neurons can be varied for the different performance of the network, while the output layer has a single neuron. An absolute mean error of 2.54% was achieved when the trained network was tested on one week’s data. This represents, on average, a high degree of accuracy in the load forecast. (
[1]
Y. Shimakura,et al.
Short-term load forecasting using an artificial neural network
,
1993,
[1993] Proceedings of the Second International Forum on Applications of Neural Networks to Power Systems.
[2]
Tomonobu Senjyu,et al.
Next Day Peak Load Forecasting Using Neural Network With Adaptive Learning Algorithm Based On Similarity
,
2000
.
[3]
S. J. Kiartzis,et al.
A neural network short term load forecasting model for the Greek power system
,
1996
.
[4]
R. Engle,et al.
Modelling peak electricity demand
,
1992
.
[5]
T. Funabashi,et al.
One-Hour-Ahead Load Forecasting Using Neural Networks
,
2002
.
[6]
C. Cañizares,et al.
ANN-based short-term load forecasting in electricity markets
,
2001,
2001 IEEE Power Engineering Society Winter Meeting. Conference Proceedings (Cat. No.01CH37194).