Title :
Short Term Load Forecasting Model in the Power System Using Ensemble of Predictors
Author :
Siwek, Krzysztof ; Osowski, Stanislaw
Author_Institution :
Warsaw Univ. of Technol., Warsaw
Abstract :
The paper present the new method of accurate prognosis of the short term load pattern for 24 hours ahead. The method uses the ensemble of neural predictors combined together by applying the blind source separation approach. Thanks to this the less accurate prognoses are not rejected but used to improve the accuracy of the final forecast. The numerical results concerning the prediction of 24-hour pattern of the load in the Polish power system have been presented and discussed in the paper.
Keywords :
blind source separation; load forecasting; power systems; Polish power system; blind source separation; forecast accuracy; neural predictors; predictor ensemble; short term load forecasting; Blind source separation; Load forecasting; Load modeling; Matrix decomposition; Neural networks; Paper technology; Power generation; Power system modeling; Predictive models; Source separation; blind source separation; ensemble of predictors; load forecast; neural networks;
Conference_Titel :
Instrumentation and Measurement Technology Conference Proceedings, 2007. IMTC 2007. IEEE
Conference_Location :
Warsaw
Print_ISBN :
1-4244-0588-2
DOI :
10.1109/IMTC.2007.379284
