In this study, one of the most complicated problems in water resources engineering, i.e., rainfall-runoff modeling is introduced and nine soft computing-based modeling approaches are considered to describe the rainfall-runoff process in a particular case study. For each of these nine approaches, many modeling choices are evaluated and the best modeling choice is selected by an intuitive two-stage competition among all modeling choices for a particular modeling approach. This competition is then applied among the best modeling choice of applied approaches and the best one is highlighted. The results shows that the modeling efficiency increases by moving toward neural modeling; particularly, the fuzzy clustering-based neural network is the most efficient and accurate paradigm among performed systems for modeling rainfall-runoff process in the considered case study. In addition, for interpreting the results, a new concept, i.e., intelligence space, and its consequent definitions are introduced that can be used as a general framework for comparing intelligent systems.

Keywords