A rolling bearing fault diagnosis method based on wind driven optimization BP neural network was proposed. BP neural network weights and threshold were the optimization parameters, wind driven algorithm was used for its optimization. The training efficiency and accuracy of the neural network were improved, the convergence rate of the network was speeded up. The vibration signal of the rolling bearing was processed to extract the characteristics of time domain, frequency domain, FFT spectrum, power spectrum and wavelet envelope spectrum as the fault feature of the bearing. The diagnosis results of the optimization algorithm are correct, and the validity and practicability of the BP neural network for the fault diagnosis of rolling bearings are verified.