Deep learning-based multilabel compound-fault diagnosis in centrifugal pumps

In this study, the issue of centrifugal pumps used in marine engineering is addressed, as they are susceptible to malfunction owing to long-term operation in highly corrosive seawater and extreme weather conditions, resulting in operational interruptions and safety risks. We propose a high-precision intelligent fault-diagnosis method for multiple fault types based on deep learning. In this method, continuous wavelet transform is firstly employed to extract signal time–frequency domain features. Subsequently, the Swin transformer model is used to process the wavelet time–frequency images converted from signals. Finally, multilabel classification methods are combined to diagnose various complex faults. The effectiveness of the proposed method is validated using a dataset obtained from simulation experiments pertaining to centrifugal-pump faults. The results show that the proposed method achieves 100% accuracy in diagnosing 27 types of faults and provides excellent diagnosis even under limited compound-fault samples, thus offering an efficient and practical method for fault diagnosis in centrifugal pumps used in marine engineering. •Highlight 1: A comprehensive method is proposed that integrates CWT, the Swin transformer model, and multilabel classification techniques for centrifugal-pump fault diagnosis.•Highlight 2: The performance of the proposed method is discussed under various extremely complex conditions and its efficiency and robustness are verified in complex mechanical fault-diagnosis tasks, particularly in cases involving data imbalance.•Highlight 3: An experimental simulation of a fault test rig for centrifugal pumps rig is designed and implemented, fault samples for model training and validation are acquired successfully, and the authenticity and reliability of the experimental data are verified via spectral analysis.