Comparative study of neural network, fuzzy logic and linear transfer function techniques in daily rainfall-runof modelling under different input domains

摘要

This paper compares artificial neural network (ANN), fuzzy logic (FL) and linear transfer function (LTF)-based approaches forndaily rainfall-runoff modelling. This study also investigates the potential of Takagi-Sugeno (TS) fuzzy model and the impact ofnantecedent soil moisture conditions in the performance of the daily rainfall-runoff models. Eleven different input vectors undernfour classes, i.e. (i) rainfall, (ii) rainfall and antecedent moisture content, (iii) rainfall and runoff and (iv) rainfall, runoff andnantecedent moisture content are considered for examining the effects of input data vector on rainfall-runoff modelling. Usingnthe rainfall-runoff data of the upper Narmada basin, Central India, a suitable modelling technique with appropriate model inputnstructure is suggested on the basis of various model performance indices. The results show that the fuzzy modelling approachnis uniformly outperforming the LTF and also always superior to the ANN-based models. Copyright  2010 John Wiley &nSons, Ltd.