Power lithium battery failure causes has a certain complexity and uncertainty．To this end,this paper proposes a fault diagnosis expert system based on fuzzy neural network,This method combines the fuzzy mathematics,the advantages of neural network and expert system．Using fuzzy mathematics can be blurred to characterize the membership degree of the fault symptoms；neural network has good self-learning ability；the expert system have strong reasoning ability；All three together,that is,to improve the accuracy of the system and operability,meet again for intelligent fault diagnosis,the requirement of automation．The test results show that the method can accurately judge the fault in the system,not only to increase the accuracy of fault detection to 0.001,the prediction error between 1%and 8%,and shorten the testing time．Improve the adaptive ability of the power lithium batteries,the independent learning ability,power lithium battery fault diagnosis is put forward a new method for scientific and efficient．