Title :
Artificial neural network short-term electrical load forecasting techniques
Author :
Xu, Leyan ; Chen, Wei Ji
Author_Institution :
Fac. of Sci. & Technol., Univ. of Macau, Macau
Abstract :
This paper presents some practical techniques for artificial neural network short-term load forecasting problems. The model described in this paper is a backpropagation based multi-layer perceptron including temperature factor. In order to expedite the training process, variable learning rate method and quasi-Newton method are employed. This paper also provides intelligent treatment to holidays and weekends in order to improve the forecasting accuracy
Keywords :
backpropagation; load forecasting; multilayer perceptrons; power system analysis computing; artificial neural network; backpropagation based multi-layer perceptron; forecasting accuracy improvement; holidays; quasi-Newton method; short-term electrical load forecasting; temperature factor; variable learning rate method; weekends; Artificial intelligence; Artificial neural networks; Cost function; Economic forecasting; Load forecasting; Multilayer perceptrons; Neurons; Power generation; Predictive models; Temperature;
Conference_Titel :
TENCON 99. Proceedings of the IEEE Region 10 Conference
Conference_Location :
Cheju Island
Print_ISBN :
0-7803-5739-6
DOI :
10.1109/TENCON.1999.818707
