The fuzzy system is used to model contact resistance.Firstly,the test is processed by the total factor method,and lots of data are obtained.The part that can reflect the characteristics of whole sample space is selected as the training data to train fuzzy system.The other part called testing data is used to evaluate the fuzzy system and verify the dependability of fuzzy system.Afterwards,according to the characteristics of basic genetic algorithm and recursive least square algorithm,the two algorithms are combined to form hybrid algorithm,where the recursive least square algorithm is processed after basic genetic algorithm.By training data,fuzzy system is trained by recursive least square algorithm,basic genetic algorithm and hybrid algorithm,and it is found that the convergence of hybrid algorithm is better than that of the other algorithms.Through using regression analysis,the model of contact resistance is found.All models are tested through testing data,the dependability is verified and the prediction result is compared.The fuzzy system obtained by hybrid algorithm has the best prediction effect.The result of the prediction and comparison shows that if the training data is enough,the fuzzy system trained by hybrid algorithm is reliable to predict contact resistance.