| Abstrakt: |
Flood routing in natural channels is essential to predict the location and the physical properties of the flood waves, which is very important for the design of hydraulic structures and flood control processes and strategies. For example, the flood routing gives accurate information on flood peak and the period of water level rise. In this context, this study employed Muskingum's graphical method along with other numerical methods [least interpolation, nonlinear regression, classical fourth-order Runge–Kutta, and improved bat algorithm (IBA)] to conduct a flood routing in the Euphrates River at Husaybah region (near the Iraqi-Syrian borders). The IBA, which is based on the chaotic search tool and it improves the population homogeneity and periodicity, was employed in this study to optimize the estimated values of the Muskingum model. The flood routing in this study was developed using a big enough hydrological data (from 1993 to 2017) from the records of the Euphrates River. The collected data were processed and analyzed using the MATLAB software (version 9.6). The results of the analysis indicated that all the numerical methods can acceptably estimate the parameters of the Muskingum's graphical method, but the IBA was the best method as it fits the data with the minimum errors. In terms of the sum of squared differences between the observed and expected discharges (SSQ) were as follows (in a descending order): Nonlinear regression (2.46), Runge–kutta method (2.38), Least squares method (2.28), Graphical method (2.21), IBA (1.36). [ABSTRACT FROM AUTHOR] |
Full text from SpringerLink