Abstract: Based on the statistical analysis of transformer winding radial deformation, main insulation capacitance test data, and low-voltage short-circuit impedance test data of Inner Mongolia Electric Power (Group) Co., Ltd. in the past 15 years, the Bayesian formula is used to evaluate the radial deformation group of transformer winding. Taking the low-voltage short-circuit impedance of the transformer winding and whether the main insulation capacitance is abnormal as the fault attribute variables, the probability of the winding radial deformation failure is calculated, and the accuracy of the judgment is significantly improved compared with the judgment based on the abnormal impedance voltage or the abnormal capacitance. Based on this method, a group assessment of more than 1200 transformers in Inner Mongolia Power Grid is carried out, and the judgment accuracy of the winding radial deformation reaches 100%.