Performance Evaluation of ANN and ANFIS Models for Estimating Garlic Crop Evapotranspiration

Abstract

Estimation of evapotranspiration (ET) is necessary in water resources management, farm irrigation scheduling, and environmental assessment. Hence, in practical hydrology, it is often necessary to reliably and consistently estimate evapotranspiration. In this study, two artificial intelligence (AI) techniques, including artificial neural network (ANN) and adaptive neuro-fuzzy inference system (ANFIS), were used to compute garlic crop water requirements. Various architectures and input combinations of the models were compared for modeling garlic crop evapotranspiration. A case study in a semiarid region located in Hamedan Province in Iran was conducted with lysimeter measurements and weather daily data, including maximum temperature, minimum temperature, maximum relative humidity, minimum relative humidity, wind speed, and solar radiation during 2008–2009. Both ANN and ANFIS models produced reasonable results. The ANN, with 6-6-1 architecture, presented a superior ability to estimate garlic crop evapotranspiration. The estimates of the ANN and ANFIS models were compared with the garlic crop evapotranspiration (