Traditional intelligent fault diagnosis methods for rolling bearings heavily depend on manual feature extraction and feature selection. For this purpose, an intelligent deep learning method, named the improved deep recurrent neural network (DRNN), is proposed in this paper. Firstly, frequency spectrum sequences are used as inputs to reduce the input size and ensure good robustness. Secondly, DRNN is constructed by the stacks of the recurrent hidden layer to automatically extract the features from the input spectrum sequences. Thirdly, an adaptive learning rate is adopted to improve the training performance of the constructed DRNN. The proposed method is verified with experimental rolling bearing data, and the results confirm that the proposed method is more effective than traditional intelligent fault diagnosis methods.

No abstract

No abstract

The effective fault diagnosis of rolling bearings is of great importance in guaranteeing the normal operation of rotating machinery. However, measured rolling bearing vibration signals are highly nonlinear and interrupted by background noise, making it hard to obtain the representative fault features. Based on this, an optimal fault diagnosis method is proposed in this paper to accurately and steadily diagnose rolling bearing faults. The proposed method primarily contains the following stages. Firstly, a gated recurrent unit and a sparse autoencoder are constructed as a novel hybrid deep learning model to directly and effectively mine the fault information of rolling bearing vibration signals. Secondly, the key parameters of the constructed model are optimized by the grey wolf optimizer algorithm to achieve better diagnosis performance. Finally, the features obtained by the constructed model are input into the classifier to get the final diagnosis results. The proposed method is validated using the experimental and practical engineering bearing data and the results confirm that the diagnosis performance of the developed method is more effective and robust than other methods.

No abstract