适应值共享遗传算法是用来解决多峰函数优化的常用手段 .本文提出了可以用来提高适应值共享遗传算法搜索能力的自适应比例变换方法 .针对复杂欺骗性问题 ,本文采用不同的指数比例变换方法进行了优化 .结果表明 ,本文提出的自适应指数比例变换方法能够迅速、稳定地找到全部全局峰 .该方法尤其适用于峰半径难以精确判定的情况 Adaptive Genetic Algorithm (GA) is a common method to solve multi-peak function optimization.In this paper, we propose an adaptive proportional transformation method that can be used to improve the search ability of adaptive genetic algorithm.For complex deceptive problems, this paper uses different exponential ratio The results show that the proposed adaptive exponential scaling method can find all the global peaks quickly and stably.It is especially suitable for the case where the peak radius is difficult to be accurately determined