Abstract: The environment of overhead transmission line tower is complex, and the premise that the distribution of actual samples is consistent with that of standard samples in traditional method is destroyed. This situation results in the low identification accuracy of abnormal vibration for single identification model under different conditions. In order to improve the identification model bias, this paper proposes an abnormal vibration identification method based on the deep transfer learning for domain adaptation. The one-dimensional convolutional neural network is are utilized to auto extract the features for the abnormal vibration signal under different working conditions. Then, the transfer learning is introduced to achieve the accurate identification of tower abnormal vibration under complex conditions. The proposed method reduces the distribution difference between complex real scenes and typical scenes through optimizing scene difference regularized loss function, and obtains an effective domain adaptation model. The experiment results show that the proposed method can obviously improve the recognition results of overhead transmission line tower abnormal vibration under complex conditions, and enhance the identification accuracy.