09-09-2016, 11:43 AM
1454052390-CHETTIHarticle.docx (Size: 402.21 KB / Downloads: 7)
1. INTRODUCTION
if we look at the problem of managing water resources, modeling the rainfall-runoff relationship is a very important point, for that hydrologists develop several models for decision support and to scale hydraulic works which protect the settlements against natural hazards. The resolution of problems related to watershed management including the management of bridges and dams, prediction of floods requires perfect knowledge of their low water levels and their floods. (Najiba Chkir, 2010).
In this context, the use of a model for calculating flow rates and water estimate on the scale of a watershed becomes necessary. It exists in the literary several empirical models to model the rainfall-runoff relationship, however, these models are still classic and usable models for cases or regions quite different. In this study, we propose the application of genetic algorithms to model the rainfall-runoff relationship and verify the performance of genetic algorithms and compare the results with the observed.
The objective of this analysis is to highlight to the end with a hydrological model based on genetic algorithms that meets all the selection criteria defined by the project partners
This objective is based on the evaluation of flood flows estimated by the hydrological model and compares its flow values with hydrographs observed at the study sites and with the classic formulas used in Algeria in order to adapt the model Algerian conditions and to check the validity of empirical formulas widely used. This application is made to study the watershed Ghardaia in southern Algeria region.