Due to the real working conditions and data acquisition equipment, the collected working data of bearings are actually limited. Meanwhile, as the rolling bearing works in the normal state at most times, it is easy to raise the imbalance problem of fault types which restricts the diagnosis accuracy and stability. To solve these problems, we present an imbalanced fault diagnosis method based on the generative adversarial network (GAN) and provide a comparative study in detail. The key idea is utilizing GAN, a kind of deep learning technique, to generate synthetic samples for minority fault class and then improve the generalization ability of the fault diagnosis model. First, this method applies fast Fourier transform to pre-process the original vibration signal and then obtains the frequency spectrum of fault samples. Second, it uses the spectrum data as the input of GAN to generate the synthetic minority samples following the data distribution of the real samples. Finally, it puts the synthetic samples into the training set and builds a stacked denoising auto encoder model for fault diagnosis. To testify the effectiveness of the proposed method, a series of comparative experiments is carried out on the CWRU bearing dataset. The results show that the proposed method can provide a better solution for imbalanced fault diagnosis on the basis of generating similar fault samples. As a comparative study, the proposed method is compared to several diagnostic methods with traditional time-frequency domain characteristics. Moreover, we also demonstrate that the proposed method outperforms three widely used sample synthesis techniques, such as random oversampling, synthetic minority oversampling technique, and the principal curve-based oversampling method in terms of diagnosis accuracy and numerical stability.

For bearing remaining useful life prediction problem, the traditional machine-learning-based methods are generally short of feature representation ability and incapable of adaptive feature extraction. Although deep-learning-based remaining useful life prediction methods proposed in recent years can effectively extract discriminative features for bearing fault, these methods tend to less consider temporal information of fault degradation process. To solve this problem, a new remaining useful life prediction approach based on deep feature representation and long short-term memory neural network is proposed in this article. First, a new criterion, named support vector data normalized correlation coefficient, is proposed to automatically divide the whole bearing life as normal state and fast degradation state. Second, deep features of bearing fault with good representation ability can be obtained from convolutional neural network by means of the marginal spectrum in Hilbert–Huang transform of raw vibration signals and health state label. Finally, by considering the temporal information of degradation process, these features are fed into a long short-term memory neural network to construct a remaining useful life prediction model. Experiments are conducted on bearing data sets of IEEE PHM Challenge 2012. The results show the significance of performance improvement of the proposed method in terms of predictive accuracy and numerical stability.

This article proposes a novel deep transfer learning-based online remaining useful life (RUL) approach for rolling bearings under unknown working condition. This approach solves the following concerns: the drift of online working condition would block data accumulation and raise bias in the prediction model, and online bearing merely has early fault data when activating RUL prediction, failing to conduct transfer learning from offline data. First, a new transfer learning-based time series recursive forecasting model is constructed to generate online RUL pseudovalues via fusing prior degradation information from offline whole-life data. With such supervised information, a new deep domain-adversarial regression network with multilevel adaptation is further built to transfer prognostic knowledge from offline data to online scenario and evaluate the RUL values of online data batch. Experimental results on the IEEE PHM Challenge 2012 bearing dataset and XJTU-SY bearing dataset validate the effectiveness of the proposed approach.

Intelligent fault-diagnosis methods using machine learning techniques like support vector machines and artificial neural networks have been widely used to distinguish bearings’ health condition. However, though these methods generally work well, they still have two potential drawbacks when facing massive fault data: (1) the feature extraction process needs prior domain knowledge, and therefore lacks a universal extraction method for various diagnosis issues, and (2) much training time is generally needed by the traditional intelligent diagnosis methods and by the newly presented deep learning methods. In this research, inspired by the feature extraction capability of auto-encoders and the high training speed of extreme learning machines (ELMs), an auto-encoder-ELM-based diagnosis method is proposed for diagnosing faults in bearings to overcome the aforementioned deficiencies. This paper performs a comparative analysis of the proposed method and some state-of-the-art methods, and the experimental results on the rolling element bearings data set show the effectiveness of the proposed method not only with adaptive mining of the discriminative fault characteristic but also at high diagnosis speed.