The impulsive components induced by bearing faults are key features for assessing gear-box bearing faults.However,because of heavy background noise and the interferences of other vibrations,it is difficult to extract ...The impulsive components induced by bearing faults are key features for assessing gear-box bearing faults.However,because of heavy background noise and the interferences of other vibrations,it is difficult to extract these impulsive components caused by faults,particularly early faults,from the measured vibration signals.To capture the high-level structure of impulsive components embedded in measured vibration signals,a dictionary learning method called shift-invariant K-means singular value decomposition(SI-K-SVD)dictionary learning is used to detect the early faults of gear-box bearings.Although SI-K-SVD is more flexible and adaptable than existing methods,the improper selection of two SI-K-SVD-related parameters,namely,the number of iterations and the pattern lengths,has an adverse influence on fault detection performance.Therefore,the sparsity of the envelope spectrum(SES)and the kurtosis of the envelope spectrum(KES)are used to select these two key parameters,respectively.SI-K-SVD with the two selected optimal parameter values,referred to as optimal parameter SI-K-SVD(OP-SI-K-SVD),is proposed to detect gear-box bearing faults.The proposed method is verified by both simulations and an experiment.Compared to the state-of-the-art methods,namely,empirical model decomposition,wavelet transform and K-SVD,OP-SI-K-SVD has better performance in diagnosing the early faults of a gear-box bearing.展开更多
In this paper,we propose a compressive sampling and reconstruction system based on the shift-invariant space associated with the fractional Gabor transform.With this system,we aim to achieve the subNyquist sampling an...In this paper,we propose a compressive sampling and reconstruction system based on the shift-invariant space associated with the fractional Gabor transform.With this system,we aim to achieve the subNyquist sampling and accurate reconstruction for chirp-like signals containing time-varying characteristics.Under the proposed scheme,we introduce the fractional Gabor transform to make a stable expansion for signals in the joint time-fractional-frequency domain.Then the compressive sampling and reconstruction system is constructed under the compressive sensing and shift-invariant space theory.We establish the reconstruction model and propose a block multiple response extension of sparse Bayesian learning algorithm to improve the reconstruction effect.The reconstruction error for the proposed system is analyzed.We show that,with considerations of noises and mismatches,the total error is bounded.The effectiveness of the proposed system is verified by numerical experiments.It is shown that our proposed system outperforms the other systems state-of-the-art.展开更多
A method that attempts to recover signal using generalized inverse theory is presented to obtain a good approximation of the signal in reconstruction space from its generalized samples. The proposed approaches differ ...A method that attempts to recover signal using generalized inverse theory is presented to obtain a good approximation of the signal in reconstruction space from its generalized samples. The proposed approaches differ with the assumptions on reconstruction space. If the reconstruction space satisfies one-to-one relationship between the samples and the reconstruction model, then we propose a method, which achieves consistent signal reconstruction. At the same time, when the number of samples is more than the number of reconstruction functions, the minimal-norm reconstruction signal can be obtained. Finally, it is demonstrated that the minimal-norm reconstruction can outperform consistent signal reconstruction in both theory and simulations for the problem.展开更多
基金Project(51875481) supported by the National Natural Science Foundation of ChinaProject(2682017CX011) supported by the Fundamental Research Foundations for the Central Universities,China+2 种基金Project(2017M623009) supported by the China Postdoctoral Science FoundationProject(2017YFB1201004) supported by the National Key Research and Development Plan for Advanced Rail Transit,ChinaProject(2019TPL_T08) supported by the Research Fund of the State Key Laboratory of Traction Power,China
文摘The impulsive components induced by bearing faults are key features for assessing gear-box bearing faults.However,because of heavy background noise and the interferences of other vibrations,it is difficult to extract these impulsive components caused by faults,particularly early faults,from the measured vibration signals.To capture the high-level structure of impulsive components embedded in measured vibration signals,a dictionary learning method called shift-invariant K-means singular value decomposition(SI-K-SVD)dictionary learning is used to detect the early faults of gear-box bearings.Although SI-K-SVD is more flexible and adaptable than existing methods,the improper selection of two SI-K-SVD-related parameters,namely,the number of iterations and the pattern lengths,has an adverse influence on fault detection performance.Therefore,the sparsity of the envelope spectrum(SES)and the kurtosis of the envelope spectrum(KES)are used to select these two key parameters,respectively.SI-K-SVD with the two selected optimal parameter values,referred to as optimal parameter SI-K-SVD(OP-SI-K-SVD),is proposed to detect gear-box bearing faults.The proposed method is verified by both simulations and an experiment.Compared to the state-of-the-art methods,namely,empirical model decomposition,wavelet transform and K-SVD,OP-SI-K-SVD has better performance in diagnosing the early faults of a gear-box bearing.
基金supported by National Natural Science Foundation of China(Grant No.61501493)。
文摘In this paper,we propose a compressive sampling and reconstruction system based on the shift-invariant space associated with the fractional Gabor transform.With this system,we aim to achieve the subNyquist sampling and accurate reconstruction for chirp-like signals containing time-varying characteristics.Under the proposed scheme,we introduce the fractional Gabor transform to make a stable expansion for signals in the joint time-fractional-frequency domain.Then the compressive sampling and reconstruction system is constructed under the compressive sensing and shift-invariant space theory.We establish the reconstruction model and propose a block multiple response extension of sparse Bayesian learning algorithm to improve the reconstruction effect.The reconstruction error for the proposed system is analyzed.We show that,with considerations of noises and mismatches,the total error is bounded.The effectiveness of the proposed system is verified by numerical experiments.It is shown that our proposed system outperforms the other systems state-of-the-art.
文摘A method that attempts to recover signal using generalized inverse theory is presented to obtain a good approximation of the signal in reconstruction space from its generalized samples. The proposed approaches differ with the assumptions on reconstruction space. If the reconstruction space satisfies one-to-one relationship between the samples and the reconstruction model, then we propose a method, which achieves consistent signal reconstruction. At the same time, when the number of samples is more than the number of reconstruction functions, the minimal-norm reconstruction signal can be obtained. Finally, it is demonstrated that the minimal-norm reconstruction can outperform consistent signal reconstruction in both theory and simulations for the problem.
