In this paper, BP neural network is used to solve the IGBT open-circuit fault diagnosis problem of the O-Z-Source inverter. Firstly, the key boost equation of O-Z source inverter is obtained through theoretical derivation. Based on the introduction of O-Z source inverter, the single and double IGBT open-circuit fault model of O-Z source inverter is developed by MATLAB/Simulink, and the output voltage of various fault cases is recorded, which is used for training and testing the fault diagnosis based on the Back Propagation (BP) neural network. Finally, the superiority and effectiveness of the proposed fault diagnosis method are verified by simulation results compared with other methods.