基于自适应策略的遗传优化设计

Genetic Optimization Design Based:A Self-adaptive Approach

早熟收敛和后期收敛速度慢是标准遗传算法(SGA)的一对主要矛盾,给算法的优化效率造成很大影响,对操作算子及其遗传参数的确定实现自适应是解决该问题的有效方法。作者根据各操作算子及其参数的特征对选择、交叉、变异算子进行基于自适应策略的遗传优化设计,使算法很好地缓解了早熟收敛和后期收敛速度慢的矛盾,从而提高了优化效率。仿真结果表明,基于自适应策略的遗传算法比标准遗传算法具有更高的解精度和优化效率。

Premature convergence and low speed of convergence in later stage is the main contradiction of Standard Genetic Algorithm （SGA）, which affects the algorithm＇s optimization efficiency. It is a feasible and efficient method to solve this problem through fixing the operators and their parameters by self-adaptive. This paper introduces the Genetic Optimization Design based on self-adaptive approach in selection, crossover and mutation operators, and solves the problem of premature convergence and low speed of convergence in later stage, and optimization efficiency is correspondingly improved. The result of the experiment indicates that self-adaptive Genetic Algorithm is more accurate and efficient than SGA.