Prediction of Stage–Discharge Relationship in Bushehr Mand River Using Hybrid Fuzzy and GMDH Methods

Understanding the Stage–Discharge relationship is of great importance in the management and planning of water resources, as well as the design of hydraulic structures, the organization of rivers, and the planning of flood warning systems. With the advancement of science and increasing the speed of computing, new methods called intelligent systems have been introduced, the use of which can be a better option for modeling. In this research, Fuzzy-GMDH, RBF and ANFIS combined intelligent methods have been used, respectively. The Fuzzy-GMDH method, which is a combination of two techniques of fuzzy logic (Fuzzy) and Group Method of Data Handling (GMDH) for flow-prediction. To study the proposed model, real and normalized data of Mand River located in Bushehr province were used and the results were compared with two methods of RBF and ANFIS neural networks. First, important parameters were determined using sensitivity analysis and then the mentioned methods were evaluated and compared with statistical indicators such as Nash coefficient (NASH). The results showed that the raw data in the Fuzzy-GMDH method with a value of NASH = 0.8822, offers higher accuracy than normal data (NASH = 0.8888). On the other hand, the Fuzzy-GMDH model has a better performance in predicting discharge daily than the other two methods with flow coefficient values of 0.8555 and 0.8776, respectively.