Parameter Estimation of a Jet Flow Model Using Genetic Algorithms and Maximum Likelihood Formulation

Abstract

The aim of this study is to develop a genetic algorithm for parameter estimation of a jet flow mathematical model, considering its axial velocity. A jet flow consists of the injection of air through a pipe or nozzle of small section. The study of these flows is useful for the design of dust separators in which this type of flows are common. Genetic algorithms are used to estimate the parameters of the velocity model, proposing different alternatives for the genetic operators. A novel selection mechanism is proposed and compared to classical methods. The parameters are also estimated by solving maximum likelihood optimization problems using a successive quadratic programming strategy. The optimal representation achieved is compared against experimental data from literature and previous estimations. The proposed genetic algorithm proves to be an efficient technique for solving the parameter estimation problem in the jet flow model.