Application of Genetic Algorithms for Estimation of Flood Routing Model Parameters

TLDR: In this paper, the results of the study of application of genetic algorithm for optimal parameter estimation of both linear and non-linear flood routing models to a case study were presented and the results had clearly depicted that the genetic algorithm is an efficient and robust means for estimation of flood routing model parameters.

Abstract: Flood routing through rivers and channels is an essential activity in hydrological analysis and this is particularly important because of the increasing emphasis that has been placed on dam-safety worldwide and due to the increasing urbanization near river channels. The routing of flood through river channels may be accomplished using two basic approaches namely hydrologic routing approach and hydraulic routing approach. There are different methods currently in usage and the Muskingum method is the most popular method and generally used by hydrologists and engineers. However, the reliability of this method is heavily depends upon the accuracy of the parameters namely K and x or C 0 , C 1 and C 2 of the model. These parameters are usually estimated by trial and error procedure. Muskingum model together with the Model proposed by Loucks (1989) have been considered for the present study and the parameters of these models were estimated using genetic algorithms, new search procedures for function optimization that apply the mechanics of natural genetics and natural selection to explore a given search space. This paper presents the results of the study of application of genetic algorithm for optimal parameter estimation of both linear and non-linear flood routing models to a case study. The sensitivity analysis of these estimated parameters was also carried out. The results had clearly depicted that the genetic algorithm is an efficient and robust means for estimation of flood routing model parameters.