In view of the frequent faults of ammeter,which seriously affect the benefit of power grid companies and the daily life of users,it is of great significance to accurately diagnose the fault types of ammeters and timely repair faulty ammeters to ensure the normal operation of power system.A diagnostic method was presented based on improved artificial bee colony algorithm to optimize the parameters of support vector machine (CABC-SVM).The advantages of ergodicity and randomness of the Tent chaotic search were used to enhance the global search capability of the swarm optimization algorithm.CABC-SVM diagnosis model was established by training the historical fault data.The results show that the classification accuracy of CABC-SVM model is up to 98.0%,and the number of convergent iterations is less than those of traditional ABC-SVM,PSO-SVM and GA-SVM.So CABC-SVM is an efficient method for ammeters fault diagnosis with higher classification accuracy and less running time.