Rainfall Runoff Modeling using Gene Expression Programming and Artificial Neural Network
Raviraj Singh1, Sunil Ajmera2

1Raviraj Singh, M.E student, civil engineering department, Shri. G.S. Institute of Technology & Science, Indore, Madhya Pradesh, India.
2Dr. Sunil Ajmera, Professor in civil engineering department, Shri. G.S. Institute of Technology & Science, Indore, Madhya Pradesh, India.

Manuscript received on February 01, 2020. | Revised Manuscript received on February 05, 2020. | Manuscript published on February 30, 2020. | PP: 978-983 | Volume-9 Issue-3, February, 2020. | Retrieval Number: B4264129219/2020©BEIESP | DOI: 10.35940/ijeat.B4264.029320
Open Access | Ethics and Policies | Cite | Mendeley
© The Authors. Blue Eyes Intelligence Engineering and Sciences Publication (BEIESP). This is an open access article under the CC BY-NC-ND license (http://creativecommons.org/licenses/by-nc-nd/4.0/)

Abstract: In water resource management and planning the Rainfall-Runoff models play a crucial role and depends mainly on the data available for planning activities. The rainfall-runoff relationship comes under the nonlinear and complex hydrological Event. In the present study two data driven modeling approaches, Artificial Neural Network (ANN) and Gene Expression Programming (GEP) has been used for modeling of rainfall-runoff process as these methods does not consider the physical nature of the process, which is complex to understand. GEP and ANN are used to model rainfall-runoff relationship for Dindori catchment in upper Narmada River Basin. Daily hydro-meteorological data of Dindori gauging station and precipitation of the catchment for a period of eighteen years were used as input in the model design. Various combinations of input variables for training and testing of models were selected based on statistical parameters. The performance of model was evaluated in term of the root mean square error (RMSE), coefficient of determination, RMSE to standard deviation ratio (RSR) and Nash Sutcliffe Efficiency. The results obtained after applying the two techniques were compared. Which indicates that GEP performed better in all performance evaluation parameters (R2 is 0.92) then ANN (R2 0.90) and is able to give mathematical relationship for rainfall-runoff modeling.
Keywords: Gene Expression Programming, Artificial Neural Network, Rainfall-Runoff.