Development of ANN, FIS and ANFIS Models to Evaluate the Adequacy Index in Agricultural Water Distribution Systems (Case study: Rudasht Irrigation Network)

In order to properly manage water in the agricultural sector, it is necessary to improve the management of agricultural water distribution systems as well as their evaluation. In this research, to achieve this goal, the models of Fuzzy Inference System (FIS), Artificial Neural Network (ANN) and Adaptive Fuzzy Neural Inference System (ANFIS), to develop a smart model for analyzing the adequacy of water delivery in an irrigation canal, given uncertainty. In order to develop and evaluate the performance of the developed models, the main canal of Rudasht Irrigation Network in Isfahan province, which is facing the problem of severe fluctuations in the inflow, was selected as the case study. The HEC-RAS hydrodynamic simulator model was used to generate the information needed to train and validate these models. The results showed that according to the MAPE index, the selected structures in ANN and ANFIS models in estimating the adequacy index of agricultural water delivery compared to FIS model have improved by 57.07% and 56.68%, respectively. Evaluation of the results showed that the developed models compared to conventional evaluation methods (hydrodynamic models and evaluation indicators) not only did not take time but also provided more accurate results, considering the uncertainty and also, ANN and ANFIS models performed better than FIS, so they can be used for other agricultural water distribution systems.