Evaluation of rainfall and discharge inputs used by Adaptive Network-based Fuzzy Inference Systems (ANFIS) in rainfall-runoff modeling.

Talei A., Chua L.H.C., Wong T.S.W, Journal of Hydrology, Vol. 391(3-4), pp. 248-262., 2010

This study investigates the effect of inputs used on event-based runoff forecasting by ANFIS. Fifteen ANFIS
models were compared, differentiated by the choice of rainfall and/or discharge inputs used. Inputs to the
ANFIS models consisted of sequential, pruned sequential, and non-sequential rainfall inputs. The overall comparison showed that the model used non-sequential rainfall timeseries and antecedent discharge performed well for forecasts upto 2 time steps ahead while the model that used a non-sequential rainfall time series alone gave superior results for forecasts greater than two time steps ahead.