I. Introduction
With the application of new transmission technologies, a large number of harmonics have appeared in the power system, and these harmonics cause vibration of electrical equipment, such as motors and transformers, and generate great noise, which greatly increases the failure rate of the equipment. For power transformers, implementing online monitoring and fault diagnosis technology can effectively reduce transformer failure rates and reduce operating costs [1]. The iron core is one of the main components of the transformer, and plays a crucial role in the electromagnetic performance and mechanical strength of the transformer. Equipment troubleshooting data reports over the years have shown that core failures often lead to a variety of transformer failures such as vibration, noise and structural loosening, etc. [2] Therefore, analyzing the vibration characteristics of the iron core is a very important part of condition monitoring of large power transformers, and has received widespread attention.