Title :
Fuzzy Markov predictor with first and second-order dependences
Author :
Teixeira, Marcelo Andrade ; Zaverucha, Gerson
Author_Institution :
COPPE, Univ. Fed. do Rio de Janeiro, Brazil
Abstract :
We present two new versions of the fuzzy Markov predictor (FMP) with different dependences among the inputs: first-order and second-order dependences. The FMP is a modification of the hidden Markov model in order to enable it to predict numerical values. The FMP can be seen as an extension of the fuzzy Bayes predictor. These hybrid systems are applied to the task of monthly electric load forecasting and successfully compared with one fuzzy system, and two traditional forecasting methods: Box-Jenkins and Winters exponential smoothing.
Keywords :
forecasting theory; fuzzy set theory; hidden Markov models; load forecasting; electric load forecasting; first-order dependences; fuzzy Bayes predictor; fuzzy Markov predictor; hidden Markov model; numerical value prediction; second-order dependences; Computer science; Fuzzy neural networks; Fuzzy systems; Hidden Markov models; Load forecasting; Niobium compounds; Power engineering and energy; Shape; Smoothing methods; Systems engineering and theory;
Conference_Titel :
Neural Networks, 2002. SBRN 2002. Proceedings. VII Brazilian Symposium on
Print_ISBN :
0-7695-1709-9
DOI :
10.1109/SBRN.2002.1181439
