Gear Fault Diagnosis Based on Kurtosis Criterion VMD and SOM Neural Network

Abstract

A gear fault diagnosis method based on kurtosis criterion variational mode decomposition (VMD) and self-organizing map (SOM) neural network is proposed. Firstly, the VMD algorithm is used to decompose the gear vibration signal, and the instantaneous frequency mean is calculated as the evaluation index, and the characteristic curve is drawn to screen out the most relevant intrinsic mode functions (IMFs) of the original vibration signal. Then, the number of VMD decompositions is determined, and the kurtosis value of IMFs are extracted to form the feature vectors. Then, the kurtosis value feature vectors of IMFs are normalized to form the kurtosis value normalized vectors. Finally, the normalized vectors of kurtosis value are input into SOM neural network to realize gear fault diagnosis. When the number of training times of SOM neural network is 100, the gear fault category is accurately classified by SOM neural network. The results show that when the training times of SOM neural network is 100 times, the gear fault diagnosis method, based on the kurtosis criterion VMD and SOM neural network is 100%, which indicates that the new method has a good effect on gear fault diagnosis. Appl. Sci. 2019, 9, 5424 2 of 25 the decomposed mode is fixed. It is better to use different wavelet bases for the analysis of different signals to achieve the best processing effect. EMD and LMD are prone to endpoint effect and mode mixing during the decomposition process. Therefore, Dragomiretskiy [9] proposed a variational mode decomposition to solve the problems caused by EMD and LMD in the decomposition process. Wang [10] used variational mode decomposition (VMD) to extract various fault features of the gearbox under strong noise environment, and compared with the ensemble empirical mode decomposition (EEMD) decomposition results, it shows that the algorithm can effectively improve the signal-to-noise ratio of the signal. Li [11] proposed a fault diagnosis method based on VMD and generalized composite multi-scale dynamic entropy (GCMSDE) to identify different health conditions of planetary gearboxes. Feng [12] uses VMD to decompose the planetary gearbox vibration signal into several intrinsic mode functions (IMFs), and performs Fourier transform on the amplitude envelope and instantaneous frequency of the sensitive IMFs to obtain the amplitude and frequency demodulation spectrum. The planetary gearbox faults have been detected based on demodulation and have been successfully identified on all three gears (sun gear, planetary gear, and ring gear). Wang [13] used the improved VMD algorithm to diagnose the gearbox and compared it with EEMD to verify the effectiveness of the proposed method. Si [14] proposes an improved VMD linked wavelet denoising method, which can suppress high frequency narrowband noise and normal noise in electromagnetic acoustic transducer (EMAT) signal, and this method can retain defect information.In recent years, researchers have studied a large number of fault classification algorithms. Among them, the gear fault...

Gear Fault Diagnosis Based on VMD Sample Entropy and Discrete Hopfield Neural Network

“…According to reference [30], the optimal decomposition number of the VMD is selected through the change in instantaneous frequency mean value. is paper also selects the optimal decomposition number of the VMD by the same method.…”

A Classification Algorithm of Fault Modes-Integrated LSSVM and PSO with Parameters’ Optimization of VMD

“…ese methods may alleviate the modal aliasing problem of high-frequency signals. Meanwhile, to optimally select the parameter K of VMD, the genetic variation sample group, kurtosis criterion variational mode decomposition, and self-organizing mapping (SOM) neural network have been adopted to adaptively determine the optimal value of the parameter K in [9][10][11][12]. To verify the effectiveness of these new methods, some experiment examples have been used to simulate in [13][14][15][16][17].…”

Locomotive Gear Fault Diagnosis Based on Wavelet Bispectrum of Motor Current

“…Driving gears in railway locomotives are used to transmit the torque on the traction motor or on the combustion engine to the wheel set, usually operating at higher speeds. us, the diagnosis of gear faults has significant importance for both safety and reliability [4][5][6][7][8][9]. Vibration signal processing is widely used to detect gear faults, but its application is limited by transducer installation [10][11][12][13][14][15].…”