Yanchao Meng

The vibration signals of rolling bearings exhibit nonlinear and non-stationary characteristics under the influence of noise. In intelligent fault diagnosis, unprocessed signals will lead to weak fault characteristics and low diagnostic accuracy. To solve the above problem, a fault diagnosis method based on parameter optimization feature mode decomposition and improved deep belief networks is proposed. The feature mode decomposition is used to decompose the vibration signals. The parameter adaptation of feature mode decomposition is implemented by improved whale optimization algorithm including Levy flight strategy and adaptive weight. The selection of activation function and parameters is crucial in the application of deep belief networks. The improved deep belief networks are proposed based on Hard swish activation function and the improved whale optimization algorithm. The optimized diagnosis model is applied to weak fault diagnosis of bearings. The experimental results show that this method has faster convergence speed and anti-noise performance under the interference of −5 db noise, and effectively improves the adaptive feature extraction ability of the model and the accuracy of fault diagnosis. Under different conditions, the average accuracy reached 97.8% and 97.6%, with good generalization performance.

The vibration signal of rolling bearing fault is nonlinear and nonstationary under the interference of background noise, and it is difficult to extract fault features from it. When feature mode decomposition is used to analyze signals, prior parameter settings can easily affect the decomposition results. Therefore, a fault feature extraction method based on improved whale optimization algorithm is proposed to optimize feature modal decomposition parameters. The improved WOA integrates Lévy flight and adaptive weight, and envelope entropy is used as fitness function to optimize feature modal decomposition parameters. The feature mode decomposition of the original signal is performed using the optimal combination of parameters to obtain multiple IMF components. The optimal IMF component envelope demodulation analysis is selected according to the kurtosis value, and the fault feature is extracted through the envelope spectrum. Comparing the LMWOA method with PSO and WOA methods by simulated and experimental signals, the results show that the optimization speed of LMWOA is faster than that of other methods. Compared with CEEMD, VMD, and FMD methods, the improved WOA-FMD method has higher fault feature ratio and can accurately extract fault features under noise interference. This method can effectively solve the parameter adaptive ability and improve the accuracy of fault diagnosis, which has practical significance.