Comparative Study of ANN and ANFIS for the Prediction of Groundwater Level of a Watershed

Fuzzy set theory, Fuzzy logic and Neural Networks techniques seem very well suited for modeling and controlling a real system. Groundwater is of major importance to civilization, because it is the largest reserve of drinkable water in regions where humans can live. The estimation of the water table elevation is one of the important aspects to understand the mechanism which comprises groundwater resources and to predict what might happen under various possible future conditions. Here, we have developed and compared two different models, Adaptive neuro-fuzzy systems (combination of fuzzy and artificial neural network systems) and Feedfoward Neural Networks systems, for the prediction of groundwater level of a watershed. Using available MATLAB software for both algorithms, the objective is to find which solution performs “better” comparing the performances of the solutions through different parameters for a specific case.