17 November 2024 in Expert Systems with Applications
6 May 2024 in Mechanical Systems and Signal Processing
22 August 2023 in Mechanical Systems and Signal Processing
6 September 2023 in IEEE Transactions on Instrumentation and Measurement
Compound faults are common in wind turbine gearbox due to complex structure and harsh operating condition, which are represented by the bearing-gear fault and gear-gear fault etc . It is a challenging task to accurately detect multiple faults from on-site vibration signal potentially contaminated by intensive background noise. To address this issue, a multi-periodicity induced sparse-fidelity representation model is proposed in this paper. The proposed method is based on the periodicity-induced overlapping group shrinkage (POGS) model with the constraint of the sparsity within and across groups (SWAG). The lp -norm and proposed reweighted minimax-concave penalty function with adjustable order are adopted for SWAG constraints. The hanning sequence is utilized as the periodicity-induced sequence in the POGS model. The closed-form solution of the proposed method is deduced using the majorization minimization algorithm. Eventually, a weighted strategy for the decomposed results is inferred to prevent over-decomposition issue. The proposed method is validated by simulation signals, experimental signal and on-site signal of actual wind turbine, which indicate that the characteristics of compound fault can be extracted plainly and the energy-loss is reduced effectively.
3 October 2024 in IEEE Transactions on Instrumentation and Measurement
When incipient faults of gears or bearings occur in wind turbine gearbox, their fault information will be readily masked by intensive background noise, which may postpone the fault detection and maintenance schedule. Specifically, the phenomenon manifested in the time domain is that the fault impulse is not obvious, while in the frequency domain, it is manifested that the fault frequency is prone to submerge in other interference frequency components. However, conventional signal processing methods and matrix factorization algorithms are difficult to intuitively characterize fault feature. Complex algorithms often require some manually adjusted parameters, which is difficult to be applied in engineering practice. To balance the performance and complexity of algorithms, a sparsity and periodicity induced symplectic geometry decomposition (SPSGD) method is proposed. The initial components are obtained by the symplectic geometry mode decomposition method. Stem from the periodicity of gear faults in time domain and the sparsity in frequency domain, a periodicity induced Bayesian clustering method is proposed as a novel iteration termination condition to pursue the fault-related components in time domain. Then, a sparsity and periodicity induced inverse filter is designed by Laplace operator to weight the faulty components for reducing the noise influence. The proposed SPSGD is validated by three cases, which indicate that the noise can be reduced adaptively as well as enhance the useful fault components effectively.
8 May 2023 in IEEE Transactions on Instrumentation and Measurement
The key steps of gram methods for bearing fault diagnosis are using an indicator to identify the optimal frequency band for demodulation and obtaining the signal’s squared envelope spectrum (SES). Using cyclostationary-based indicators for optimal band determination is one of the most widely used methods. However, the effectiveness of these methods may be compromised when the signal is non-Gaussian distributed. Compared with cyclostationary analysis, correntropy has the potential capability to detect cyclostationarity under non-Gaussian conditions. Therefore, this study proposes a novel tool called the spectral correntropy-based coherence gram (IEESCCgram) as an optimal band determination and fault diagnosis method. First, a novel bi-frequency map, the spectral correntropy-based coherence (SCECoh), is proposed, which is more robust to impulsive noise and can reveal hidden cyclic periodicities. Then, by represented in a 1/3 binary tree, the IEESCCgram uses the proposed cyclic band spectral negentropy (SNc) to select the optimal band. Finally, the improved energy spectrum (IEES) is obtained for bearing fault diagnosis. Vibration signals collected from experimental test rigs and real-life applications are used to verify the effectiveness and robustness of the IEESCCgram for bearing fault diagnosis when the signal is contaminated by impulsive noise and cyclostationary interference.