The accuracy of the intelligent diagnosis of rolling bearings depends on the quality of its vibration data and the accuracy of the state identification model constructed accordingly. Aiming at the problem of “poor quality” of data and “difficult to select” structural parameters of the identification model, a method is proposed to integrate data cleaning in order to select effective learning samples and optimize the selection of the structural parameters of the deep belief network (DBN) model. First, by calculating the relative energy fluctuation value of the finite number of intrinsic function components using the variational modal decomposition of the rolling bearing vibration data, the proportion of each component containing the fault component is characterized. Then, high-quality learning samples are obtained through screening and reconstruction to achieve the effective cleaning of vibration data. Second, the improved particle swarm algorithm (IPSO) is used to optimize the number of nodes in each hidden layer of the DBN model in order to obtain the optimal structural parameters of the intelligent diagnosis model. Finally, the high-quality learning samples obtained from data cleaning are used as input to construct an intelligent identification model for rolling bearing faults. The results showed that the proposed method not only screens out the intrinsic mode function components that contain the fault effective components in the rolling bearing vibration data, but also finds the optimal solution for the number of nodes in the DBN hidden layer, which improves bearing state identification accuracy by 3%.

The vibration signal of rolling bearing exhibits the characteristics of energy attenuation and complex time-varying modulation caused by the transmission with multiple interfaces and complex paths. In view of this, strong ambient noise easily masks faulty signs of rolling bearings, resulting in inaccurate identification or even totally missing the real fault frequencies. To overcome this problem, we propose a reinforced ensemble local mean decomposition method to capture and screen the essential faulty frequencies of rolling bearing, further boosting fault diagnosis accuracy. Firstly, the vibration signal is decomposed into a series of preliminary features through ensemble local mean decomposition, and then the frequency components above the average level are energy-enhanced. In this way, principal frequency components related to rolling bearing failure can be identified with the fast spectral kurtosis algorithm. Finally, the efficacy of the proposed approach is verified through both a benchmark case and a practical platform. The results show that the selected fault characteristic components are accurate, and the identification and diagnosis of rolling bearing status are improved. Especially for the signals with strong noise, the proposed method still could accurately diagnose fault frequency.