CGRG Bibliography of Canadian Geomorphology
Search Results
Author : Coulibaly, P.; and Baldwin, C.K.
Date : 2002.
Title : Nonstationary hydrologic time series forecasting using dynamic recurrent neural networks.
Publication : Annual Meeting of the Canadian Geophysical Union. May 18-21, 2002, Banff, Alberta.
Issue :
Page(s) :
Abstract
Evidence of nonstationary trends in hydrological time series as result of natural and / or anthropogenic climate variability has raised more interest in nonlinear dynamic system modelling methods. Here, we investigate the effectiveness of dynamically driven recurrent neural networks (RNN) for complex time-varying hydrological system modelling. A dynamic RNN approach is proposed to directly forecast different nonstationary hydrologic time series. The proposed method automatically selects the most optimally trained network in any case. The simulation performance of the dynamic RNN-based model is compared with the results obtained from optimal multivariate adaptive regression splines (MARS) models. It is shown that the dynamically driven RNN model can be a good alternative for the modelling of complex dynamics of hydrological system, performing better than MARS model on the two selected time series.
Bibliography of Canadian Geomorphology