Title :
Multiple classifier system for short term load forecast of Microgrid
Author :
Chan, Patrick P K ; Chen, Wei-chun ; Ng, Wing W Y ; Yeung, Daniel S.
Author_Institution :
Machine Learning & Cybern. Res. Center, South China Univ. of Technol., Guangzhou, China
Abstract :
During last decade, Microgrid has been an area of intense study. It also becomes more important in Smart Grid (SG). Short-term load forecast (STLF) in Microgrid is an important factor for planning and optimization of distributed generation. However, STLF for Microgrid is a tough and complex assignment because load of a Microgrid could be change rapidly within a short period of time. The present work proposes on-line learning model of Microgrid short-term load forecast using Multiple Classifier Systems (MCSs). This model is constructed from different training sets and dynamic weighting fusion is used. The proposed method is evaluated by the real Microgrid dataset in Hong Kong comparing with other existing methods experimentally.
Keywords :
distributed power generation; learning (artificial intelligence); load forecasting; power distribution planning; power engineering computing; smart power grids; Hong Kong; MCS; SG; STLF; distributed generation optimization; distributed generation planning; dynamic weighting fusion; microgrid; multiple-classifier system; on-line learning model; short-term load forecast; smart grid; training sets; Cybernetics; Distributed power generation; Electricity; Forecasting; Load forecasting; Machine learning; Microgrid; Multiple classifier system; Short-term load forecast (STLF);
Conference_Titel :
Machine Learning and Cybernetics (ICMLC), 2011 International Conference on
Conference_Location :
Guilin
Print_ISBN :
978-1-4577-0305-8
DOI :
10.1109/ICMLC.2011.6016936
