Redes neurais, metodologias de agrupamento e combinação de previsores aplicados a previsão de vazões naturais

AUTOR(ES)
DATA DE PUBLICAÇÃO

2004

RESUMO

In addition, this work suggests a linear approach to combine forecasts generated by a set of individual forecasting models in a simple and effective way. We use, as a combiner, a neural network trained with the gradient descent algorithm. The aim is to combine the forecasts generated by the different forecasting models as an attempt to capture the contributions of the most important prediction features of each individual model at each prediction step. The approach is also used for streamflow time series prediction choosing, as individual forecasting models, the most promising predictive methods. Experimental results with actual data suggest that the predictive clustering approach performs globally better than the current streamflow forecasting methodology adopted by many hydroelectric systems worldwide, and a fuzzy neural network, a nonlinear prediction model. The combination approach, with lower prediction errors, performs better than each of the individual forecasting models.

ASSUNTO(S)

sistemas difusos assunto em ingles analise de series temporais previsão hidrologica redes neurais (computação)

Documentos Relacionados