Publication

Application of Adaptive Network-based Fuzzy Inference Systems in Event-based Rainfall-Runoff Modeling in a tropical catchment

Nadeem Nawaz, Sobri Harun, Amin Talei, Tak Kwin Chang, International Conference on Water Resources, 24-25 Nov 2015, Langkawi, Malaysia, 2015

Urbanization as a result of development in fast developing countries such as Malaysia leads to gradual transformation of agricultural or forest landscapes to built-up areas. Such change has significant impact on hydrologic process in the catchment which in turn may end up with an increase in both magnitude and frequency of floods in urban areas. Therefore, reliable rainfall-runoff models that are able to estimate discharge of a catchment accurately are in need. To date, several physically-based models are developed to capture the rainfall-runoff process; however, they require significant number of parameters which could be difficult to be measured or estimated. Recently, neuro-fuzzy systems (NFS) which are well-known for their ability in simulating complex nonlinear systems have been widely used in hydrological time series modeling and prediction. These models are able to identify a direct mapping between inputs and outputs with less number of physical parameters. Adaptive network-based fuzzy inference system (ANFIS) is one of the NFS that is well-practiced in hydrological modeling. In this study, 70 rainfall-runoff events are extracted from twelve years hourly rainfall and runoff data of Semenyih River catchment located in Selangor, Malaysia where 50 of them are chosen for training and the remaining 20 for testing. The study catchment has 4 rainfall stations and one discharge station at the outlet of the catchment. An input selection method based on correlation analysis and mutual information was developed to identify the proper input combinations of rainfall and discharge antecedents. The desired output was discharge at present time Q(t). The results obtained by ANFIS model were then compared with an autoregressive model with exogenous inputs (ARX) as a bench mark. Results showed that ANFIS outperforms ARX model and has a good potential to be used as a reliable rainfall-runoff modeling tool.