Swarm intelligence approach for parametric learning of a nonlinear river flood routing model

Abstract

Flood routing models are mathematical methods used to predict the changes over the time in variables such as the magnitude, speed and shape of a flood wave when water moves in a river, a stream or a reservoir. These techniques are widely used in water engineering for flood prediction and many other applications such as dam design, geographic and urban planning, disaster prevention, and so on. Flood routing models typically depend on some parameters that must be estimated from data. Several techniques have been described in the literature for this task. Among them, those based on swarm intelligence are getting increasing attention from the scientific community during the last few years. In this context, the present contribution applies a powerful swarm intelligence technique called bat algorithm to perform parametric learning of a hydrological model for nonlinear river flood routing. The method is applied to data of a real-world example of a river reach with very good results.