Coevolutionary augmented Lagrangian methods for constrained optimization

This paper introduces a coevolutionary method developed for solving constra ined optimization problems, This algorithm is based on the evolution of two populations with opposite objectives to solve saddle-point problems, The a ugmented Lagrangian approach is taken to transform a constrained optimizati on problem to a zero-sum game with the saddle-point solution. The populatio ns of the parameter vector and the multiplier vector approximate the zero-s um game by a static matrix game, in which the fitness of individuals is det ermined according to the security strategy of each population group. Select ion, recombination, and mutation are done by using the evolutionary mechani sm of conventional evolutionary algorithms such as evolution strategies, ev olutionary programming, and genetic algorithms. Four benchmark problems are solved to demonstrate that the proposed coevolutionary method provides con sistent solutions with better numerical accuracy than other evolutionary me thods.