Model Development for Estimation of Sediment Removal Efficiency of Settling Basins Using Group Methods of Data Handling

Abstract

Settling basins are generally used as sediment removal structures in which flow velocity is reduced, resulting in surplus settlement of sediment particles. The accuracy of the available empirical equations for sediment removal efficiency is checked using data available in the literature. The existing relationships of removal efficiency were not found to yield satisfactory results. Therefore, the data were reanalyzed and new models were developed using nonlinear regression and the group method of data handling (GMDH). On the basis of various performance parameters, it was observed that the proposed nonlinear regression model had the highest accuracy compared with other available relationships. It was also observed that the efficiency estimated using the GMDH model (R=0.960, AAD=14.698, and RMSE=0.094) was more accurate than that estimated by the regression model (R=0.863, AAD=30.675, and RMSE=0.171). Sensitivity analysis indicated that the ratio of particle fall velocity to average flow velocity in the basin was the most effective parameter for sediment removal efficiency. This work will help in quantifying and subsequently improving the management of surplus sediment in canals.