A Data Enhancement and Fault Diagnosis Method for Planetary Gearboxes With Extremely Small Samples

Generative adversarial networks (GANs) face challenges in effectively utilizing deep features of varying resolutions and adequately fusing information characterized by typicality and multiplicity, particularly when confronted with extremely limited training samples. A multiscale and local feature fusion-based GAN (FPN_LoFGAN) is proposed to address these issues. Initially, time-domain waveforms of vibration signals are transformed into time-frequency maps using the continuous wavelet transform (CWT). Subsequently, feature maps are extracted through channel attention (CA) mechanisms and fused with multiscale features via a feature pyramid network (FPN). Finally, adversarial samples are generated using a multihead attention-embedded generator, while a self-attention-embedded discriminator ensures robust adversarial learning. The proposed method, applied to fault analysis and diagnosis on various planetary gearbox datasets, outperforms state-of-the-art methods, achieving superior performance regarding SSIM and FID metrics.