Fault Diagnosis for Rolling Bearing of Swashplate Based on Multi-noise Data Training CNN

In order to solve the problem that the poor fault diagnosis of the helicopter swashplate rolling bearing in noisy environment by using traditional convolutional neural network (CNN), we proposed a fault diagnosis method based on multi-noise data training CNN. Firstly, different sizes of the white gaussian noises are randomly added to the vibration signals. Then, the time-frequency diagram of the vibration signals in different noise states are constructed by using the wavelet transform method. Finally, the time-frequency map under different noises is classified by CNN. Diagnostic experiments were carried with the bearing fault data of the research team and Case Western Reserve University, the results show that compared with the traditional CNN, the proposed method has higher fault recognition rate in noise environment.