模糊系统的模糊推理方法常依赖于一系列由主观决定的参数，如模糊隶属函数、模糊关系矩阵等．如何确定这些参数会直接影响系统的性能．为了能对这些参数进行优化，文中采用一种新型的模糊推理方法，在此基础上，利用遗传算法产生出模糊推理方法中的最优参数．同时，在进化演变的搜索过程中，使用不断调整适应函数的手段，解决了遗传算法过早收敛于次优解的问题，提高了遗传算法的搜索精度． The fuzzy reasoning method of fuzzy system often depends on a series of subjectively determined parameters, such as fuzzy membership function, fuzzy relation matrix and so on. How to determine these parameters will directly affect the performance of the system. In order to optimize these parameters, a new type of fuzzy inference method is adopted in this paper. Based on this, genetic algorithm is used to generate the optimal parameters in fuzzy inference method. In the meantime, in the process of evolutionary evolution, the method of continuously adjusting the adaptation function is used to solve the problem that the genetic algorithm prematurely converges to the second-best solution and improve the search precision of the genetic algorithm.