Data imbalances present a serious problem for intelligent fault diagnosis. They can lead to reduced diagnostic precision, which can jeopardize equipment reliability and safety. Based on that, this paper proposes a novel fault diagnosis method combining the denoising diffusion implicit model (DDIM) with a new convolutional neural network framework. First, the Gramian angular difference field (GADF) is used to generate 2D images, which are then augmented using DDIM. Next, by utilizing the weight-sharing properties of a convolutional neural network and the self-attention mechanism along with the global data processing capabilities of Transformers, a TransNet model is constructed. The augmented data are input into the model for training to establish a fault diagnosis framework. Finally, the method is validated and analyzed using the CWRU bearing dataset and the Nanchang Railway Bureau dataset. The results show that the proposed method achieves over 99% recognition accuracy on the two datasets. Meanwhile, the proposed model provides better generalization performance and recognition accuracy than existing fault diagnosis methods.

In the field of intelligent fault diagnosis, data imbalance significantly degrades diagnostic performance and has emerged as a major challenge. Generative adversarial networks (GANs) are commonly used data augmentation methods, but their training process is unstable and prone to mode collapse. Therefore, this paper proposes a vibration data augmentation method based on an improved residual U-Net diffusion model to avoid the mode collapse issues faced by GANs. The proposed framework extends the traditional U-Net by integrating residual blocks, which stabilize gradient propagation in deep architectures, mitigate gradient vanishing, and enhance the network’s ability to capture localized disturbances and fine-grained patterns in vibration signals. The hierarchical global attention mechanism further enables the model to emphasize local details in shallow layers while effectively constructing long-range dependencies and global structures in deeper layers. In addition, the incorporated channel-compression strategy substantially reduces the computational overhead of self-attention without compromising representational capacity. The diffusion time-step conditional embeddings guide the model to adopt stage-specific reconstruction strategies under varying noise intensities. This progressive denoising process—from coarse structural recovery to refined detail enhancement—significantly improves the fidelity and overall quality of the generated signals. Then, a diffusion model based on the improved residual U-Net is constructed, leveraging the characteristics of progressive diffusion and random sampling to enhance the realism and diversity of sample generation. Furthermore, to fill the gap of a unified evaluation framework for vibration data generation models, a time–frequency joint maximum mean discrepancy metric is proposed to assess the quality and performance of augmented data more effectively. Finally, the effectiveness and superiority of the proposed method are validated on both publicly available benchmark datasets and real-world engineering datasets.

For the sake of addressing the issue of extracting multiple features embedded in a noise-heavy vibration signal for bearing compound fault diagnosis, a novel model based on improved adaptive chirp mode decomposition (IACMD) and sparse representation, namely IACMDSR, is developed in this paper. Firstly, the IACMD is employed to simultaneously separate the distinct fault types and extract multiple resonance frequencies induced by them. Next, an adaptive bilateral wavelet hyper-dictionary that digs deeper into the periodicity and waveform characteristics exhibited by the real fault impulse response is constructed to identify and reconstruct each type of fault-induced feature with the help of the orthogonal matching pursuit (OMP) algorithm. Finally, the fault characteristic frequency can be detected via an envelope demodulation analysis of the reconstructed signal. A simulation and two sets of experimental results confirm that the developed IACMDSR model is a powerful and versatile tool and consistently outperforms the leading MCKDSR and MCKDMWF models. Furthermore, the developed model has satisfactory capability in practical applications because the IACMD has no requirement for the input number of the signal components and the adaptive bilateral wavelet is powerfully matched to the real fault-induced impulse response.

The incipient faults of rolling bearings are dynamically propagating in the service status. However, the bearing material and the size of the fault will affect its expansion trend and direction. Furthermore, bearings manufactured from different materials behave differently when they fail. Therefore, the influence of the fault degree on the dynamic characteristics and crack propagation of rolling bearings is investigated in this paper. First, a dynamic model of the bearing, both under fault-free conditions and with varying fault sizes on the outer ring, is established by considering the actual working conditions of rolling bearings. Then, the reliability of the dynamic model is verified theoretically and experimentally. Finally, the study examined the slip behavior of rolling elements, the variation trends in the maximum shear stress and principal stresses on the outer ring, and the direction of crack propagation under different fault severities. The results indicate that (1) the severity of roller slip becomes more pronounced with the expansion of the fault size; (2) material differences will affect the timing of macro-slip during faults; (3) crack propagation tends to initiate at the edge of the fault exit, with the propagation rate increasing as fault severity escalates; and (4) tensile stress was observed in the first principal stress, which accelerates crack bifurcation at the faulted edge, while both the second principal stress and third principal stress exhibit compressive stress, playing a suppressive role in crack bifurcation at the faulted edge. These findings provide a theoretical basis for further research on the evolution of faults in rolling bearings.