Abstract: Through the analysis of monitoring data, the fault diagnosis technology of dry type transformer can extract signal feature, fault diagnosis, state assessment and life prediction, improve the safety of dry type transformer in underground operation, and reduce the loss caused by the fault. This paper introduces the research status of signal feature extraction method (frequency response method, wavelet transform method, stacked autoencoder), fault diagnosis method (BP neural network, support vector machine, Bayesian network), condition assessment and life prediction method (cross entropy combination prediction method, grey theory, analytic hierarchy process), and compares several methods. Finally, the dry type transformer fault diagnosis technology is prospected.