Under strong background noise, the weak incipient fault features of mechanical equipment are easily obscured by environmental noise, thereby hindering effective feature extraction. Inspired by the decomposition mode of wavelet packet transform, a multiscale recursive mode decomposition (MRMD) method is proposed based on a tree-structured decomposition framework in this paper. The MRMD constructs a multi-level recursive decomposition framework that hierarchically partitions the signal across different scales, thereby enhancing the adaptivity of the decomposition process. Moreover, by employing a tree-based binary-splitting strategy, the method reduces the multi-parameter optimization problem inherent in conventional approaches to a single dominant tuning parameter focused on filter length, which substantially improves computational efficiency and parameter stability. To ensure that selected components preserve high fault-information density and periodic characteristics, the harmonic energy to background is used as the component evaluation criterion and is supplemented by a harmonic-matching verification mechanism to quantitatively identify and screen candidate components. Simulation and experimental results indicate that MRMD reliably identifies fault features across a range of signal-to-noise ratios while maintaining high diagnostic accuracy and markedly improving decomposition efficiency and adaptivity.
Haiyu Li, Jian Cheng, Jinde Zheng et al.· Measurement science and tech...· 0 citations
Although intelligent diagnosis methods based on deep learning have achieved significant theoretical advancements, the application in real-world industrial scenarios remains challenging. The scarcity of fault samples in real-world environments hinders the effective training of deep learning models, thereby compromising their diagnostic accuracy and generalization ability. To address this issue, a small sample transfer diagnosis method driven by a high-precision dynamic model for rolling bearings is proposed this paper. Firstly, a high-precision dynamic model of rolling bearing with defects is constructed to simulate vibration signals under various fault types and severities, generating a large-scale and diverse library of simulated fault samples. Then, an improved Transformer-based deep transfer learning network is developed to extract features from both simulated samples and measured small samples at local and global scales, and perform multi-layer deep domain adaptation to minimize the distribution discrepancy between the simulated and measured data, facilitating accurate fault diagnosis under small sample conditions. Finally, experimental verification on two datasets with different tasks demonstrated that the proposed method not only effectively diagnoses rolling bearing faults but also exhibits excellent generalization ability on small sample datasets.
Jinyu Tong, Guotao Chen, Xun Huang et al.· Structural Health Monitoring· 0 citations
We use cookies to run the site and, with your consent, for analytics and to show ads.
See our Cookie Policy.
