针对遗传算法参数多，且配置困难的问题，提出一种利用正交试验优化选择参数的方法，它使得对于不同领域的优化问题只需用正交试验进行一次参数配置，然后用遗传算法进行具体寻优即可取得较好效果．这种方法易于编程的实现，且在一定程度上避免了遗传算法参数配置的盲目性． Aiming at the problems that there are many parameters of GA and the configuration is difficult, this paper proposes a method of optimizing selection parameters by using orthogonal test, which makes it possible to optimize parameters in different fields by using orthogonal test only once, and then use genetic algorithm Search for better results can be achieved. This method is easy to program to achieve, and to some extent to avoid the blindness of genetic algorithm parameter configuration.