Abstract: Aiming at the problems of poor binary classification performance inherent in a single SVM and low classification accuracy of multiple classifiers using the same parameter, a transformer fault diagnosis method based on fuzzy C-mean clustering and DBO-LSSVM is proposed. Firstly, the fuzzy C-mean clustering method is used to cluster the samples, and a complete binary tree structure is constructed, with each leaf node adopting an LSSVM classifier; secondly, the dung beetle optimization algorithm(DBO) is used to optimize the kernel parameter and the penalty coefficient of each LSSVM classifier; finally, the optimal parameter is used to carry out the fault diagnosis in the complete binary tree layer by layer from the top to the bottom and analyzed and compared with the different algorithms. Simulation results show that the proposed method has high diagnostic accuracy in transformer fault diagnosis.