The modal parameter of transformer winding, as direct reflections of mechanical performance, is an important theoretical foundation in the field of transformer manufacturing and detection of winding condition based on vibration. To identify the modal parameters accurately, a modal experiment on a 10 kV transformer winding was conducted in this paper. A wavelet transforms method based on complex Morlet wavelet to identify winding natural frequency and damping ratio was proposed using the experiment to identify transformer winding modal parameters. Meanwhile, modal shapes were extracted using the least square method. The first four order modal parameters, including natural frequency, damping ratio and modal shapes were extracted from the vibration signal. The results demonstrate that all the natural frequencies are far away from the exciting frequency, and the modal shapes are symmetrical, which indicates that the design is appropriate. The parameters extracted by the general frequencydomain method PolyMax and the proposed method match well, which verifies the effectiveness of the proposed method in identifying modal parameters of transformer winding.