Haploid and diploid algorithms, a new approach for global optimization: compared performances

Abstract

In this paper, a new approach for optimization is presented. This approach tends to imitate principles of organic evolution processes as rules for an optimization procedure. The two techniques used, haploid and diploid algorithms, are based on different organic structure modelling approaches and both use genetic concepts such as population, recombination and mutation as evolution rules to guide the optimum search. The theoretical analysis of this stochastic approach, especially with regard to issues about the global convergence, is currently in progress. However, previous tests and those presented here validated the method. These genetic algorithms are presented and their performances are compared. The comparison pertains to the convergence characterized by the number of generations needed to reach the solution, and running times are also compared with respect to applications that limit search time severely. Several complex optimization problems involving concrete applications were studied with different tuning parameters. Examples were selected to show the behaviour of the two algorithms when faced with several optima and nonlinear constraints. A practical application deals with a hydrodynamical problem, where discrete and actual variables are simultaneously handled in the same parameter optimization problem.