Electrical,thermal,mechanical,and other stresses acting together on dry-type transformers may lead to core loosenings,winding deformations,and other faults. The vibration-based analysis method can sensitively reflect the mechanical state of the core and windings. Thus,it is suitable for the mechanical fault detection of dry-type transformers. In this paper,a finite element model of the dry-type transformer is constructed,and its vibration mechanism and characteristics are investigated by simulation. The simulation results show that,at normal operation,the winding vibration is much smaller than the core vibration caused by the magnetostrictive effect. A dry-type transformer experiment platform is built,and then the vibration signals of the transformer are measured and analyzed at normal and mechanical fault states. Experimental results show that with the gradual magnetic saturation of the iron core,the vibration signal is no longer linearly related to the squared voltage. Moreover,the peak value of the total vibration signal grows rapidly and the amplitude of the fundamental frequency vibration signal increase slowly. The vibration signal spectrum of the dry-type transformer under different operating conditions is measured. Experimental results show that the ratio of fundamental frequency to total and high frequency to low frequency can be used to diagnose the mechanical state of dry-type transformers.