At present, the fault diagnosis methods for rolling bearings are all based on research with fewer fault categories, without considering the problem of multiple faults. In practical applications, the coexistence of multiple operating conditions and faults can lead to an increase in classification difficulty and a decrease in diagnostic accuracy. To solve this problem, a fault diagnosis method based on an improved convolution neural network is proposed. The convolution neural network adopts a simple structure of three-layer convolution. The average pooling layer is used to replace the common maximum pooling layer, and the global average pooling layer is used to replace the full connection layer. The BN layer is used to optimize the model. The collected multi-class signals are used as the input of the model, and the improved convolution neural network is used for fault identification and classification of the input signals. The experimental data of XJTU-SY and Paderborn University show that the method proposed in this paper has a good effect on the multi-classification of bearing faults.

A voiceprint signal as a non-contact test medium has a broad application prospect in power-transformer operation condition monitoring. Due to the high imbalance in the number of fault samples, when training the classification model, the classifier is prone to bias to the fault category with a large number of samples, resulting in poor prediction performance of other fault samples, and affecting the generalization performance of the classification system. To solve this problem, a method of power-transformer fault voiceprint signal diagnosis based on Mixup data enhancement and a convolution neural network (CNN) is proposed. First, the parallel Mel filter is used to reduce the dimension of the fault voiceprint signal to obtain the Mel time spectrum. Then, the Mixup data enhancement algorithm is used to reorganize the generated small number of samples, effectively expanding the number of samples. Finally, CNN is used to classify and identify the transformer fault types. The diagnosis accuracy of this method for a typical unbalanced fault of a power transformer can reach 99%, which is superior to other similar algorithms. The results show that this method can effectively improve the generalization ability of the model and has good classification performance.

Kurtogram is a typical fault diagnosis algorithm, and its derivative methods are widely used in bearing fault signal processing. The key to such methods is to select the indexes that are specifically sensitive to periodic fault characteristics. The traditional time domain kurtosis index is too sensitive to non-periodic transient impulses, and the frequency domain index is too sensitive to harmonic components, which seriously affects the selection of resonance frequency band and the extraction of fault features. In view of this, a polynomial detrended autocorrelation kurtosis of 1.5D spectrum is proposed in this paper. Firstly, the original signal is decomposed by wavelet packet decomposition, and the 1.5D spectrum of each subband signal is calculated. Then, the 1.5D spectral autocorrelation coefficient of each subband is calculated, and the trend term is removed from the specific interval of autocorrelation coefficient by polynomial fitting. Finally, the kurtosis of the autocorrelation coefficient after removing the trend term is calculated to select the optimal resonance frequency band for analysis, so as to extract the fault feature information. The experimental data show that this method is sensitive to the fault characteristic frequency and frequency doubling with coupling relationship, and can suppress non-periodic transient impulse and harmonic interference.

No abstract

Wheelset bearing is a typical vulnerable structural component in high-speed trains and heavy haul vehicles. In addition to the typical nonlinear and nonstationary characteristics, the vibration signal of wheelset bearing also contains track subgrade vibration and transmission path coupling interference components. To solve this problem, this paper proposes a new feature extraction method for wheelset bearing faults. This method constructs the Teager energy spectrum correlation kurtosis, which is purposely sensitive to periodic fault impulse components, as the objective function. The Q-factor and redundancy of tunable Q-factor wavelet transform are selected by using the parameter convex optimization method, which makes the signal decomposition have better sparsity, so as to extract fault information accurately. Simulated analysis, experimental signal analysis of QPZZ-II test-bed, and experimental signal analysis of wheelset bearing test-bed show that the proposed method can suppress the influence of nonperiodic transient impulse components, harmonic components, and noise components in the signal and accurately extract the periodic impact characteristics of bearings.