Suspended load estimation using a MARS-Fuzzy regression model

The problem of estimation of suspended load carried by a river flow is an important topic for many water resources projects. Conventional methods are based on the assumption of exact observations. But, in practice, a major source of natural uncertainty is due to imprecise measurements and/or imprecise relationships between variables. In this paper, by using Multivariate Adaptive Regression Splines (MARS) technique, a novel fuzzy regression model is introduced for imprecise response and crisp explanatory variables. The proposed fuzzy regression model is applied to forecast suspended load by discharge using a real hydrology data set. The estimation accuracy of the proposed method is compared with two current parametric fuzzy regression models, namely, fuzzy least-absolutes model and fuzzy least-squares model, indicating the advantages of the proposed model

Suspended load estimation using a MARS-Fuzzy regression model Keywords:

Discharge , Hydrology , Multivariate Adaptive Regression Splines (MARS) , suspended load