Aiming at the transmission line fault caused by ice damage, the annual failure rate of regional lines is studied and analyzed. Firstly, Pearson-III distribution suitable for Climatic Extreme value probability analysis is used to fit the annual ice extreme value, and the distribution of ice extreme value is obtained. Based on the theory of metal deformation, the probability model of line failure under condition of icing thickness is constructed then. Finally, the annual fault rate distribution of regional lines is synthesized, and genetic algorithm is used to estimate the parameters. The results show that the Pearson-III distribution is suitable for the fitting of icing extreme value, the probability model based on metal deformation theory can truly reflect the possibility of icing fault, and the annual failure rate distribution of regional lines can be used to analyze the icing fault situation of regional lines.