混合离散变量优化的精英多父体杂交算法

Abstract

A new approach to handle constrained function optimization problems by using evolutionary algorithms is presented.The value adopting problem of hybrid discrete variable in optimization design is dealing with reasonably by the method.On the basis of Guo Tao algorithm,a elite multi-parent crossover evolutionary optimization algorithm is constructed by introducing the elite preservation strategy,constructing dynamic penalty function and enhancing the selection pressure of parents in the process of algorithm convergence.The elite multi-parent crossover evolutionary optimization algorithm program DEMPCOA1.0 with hybrid discrete variables is developed.The living example of mechanical optimization design show that this algorithm has no special requirement on the characteristic of optimal design problem.The universal adaptability is fairly good,operation of program is reliable and the ability of overall convergence is strong.