Transformer internal fault diagnosis usually requires analysis of dissolved gas in oil, but the information extraction, detection and analysis process is cumbersome, and the real-time performance is poor. Therefore, this paper proposes a fast diagnosis method of transformer internal fault which only needs electrical quantity information. The frequency domain fault characteristics of short-circuit current and differential current are extracted by wavelet packet analysis. The fault characteristics of zero-sequence current are represented by the maximum value. The information fusion technology is used to fuse all the fault features, and the BP neural network algorithm is used to diagnose the internal electrical fault types of transformer. The simulation model is established on Matlab/Simulink platform and the example analysis is carried out, the results show that the proposed internal electrical fault diagnosis method of transformer has high accuracy and high reliability.