Recent works suggest that using knowledge transfer strategies to tackle cross-domain diagnosis problems is promising for achieving engineering diagnosis. This paper presents a diagnosis scheme for rolling bearing under a challenging domain generalization scenario, in which more potential discrepancies among multiple source domains should be eliminated and only normal samples of the target domain are available during the training stage. In order to achieve sufficient generalization performance, a diagnosis scheme combined some a priori diagnosis knowledge and a deep domain generalization network for fault diagnosis (DDGFD) is elaborated. Through the signal preprocessing steps guided by the a priori diagnosis knowledge, the inputs of DDGFD with a primary consistent meaning across domains are constructed from the vibration signal. On this basis, DDGFD would intently release its talent on learning discriminative and domain-invariant fault features from source domains, and then generalize the learned knowledge to identify unseen target samples. On cross-domain tasks organized using broad bearing datasets, the superiority of DDGFD is validated by comparing its performance with various data-driven diagnosis methods.

In recent years, Convolutional neural networks (CNNs) have achieved start-of-art performance in the fault diagnosis field. If there is no available information on the unseen operating conditions, the model trained on the seen operating condition cannot perform well. One of the feasible strategies is to enhance the generalization ability of the network on various seen operating conditions. We introduce the center loss to the traditional CNN and build an end-to-end fault diagnosis framework (called CNN-C). By minimizing the intra-class variations, center loss cluster the learned features across various seen operating conditions. With the joint supervision of the center loss and the softmax loss, the learned features of the same class could minimize the domain difference across various seen operating conditions while the features of different classes are separable. The generalization ability of network is improved on unseen operating conditions. Compared with the shallow methods and traditional CNN, the proposed method is promising to deal with the fault diagnosis tasks of the bearing and gearbox.

a novel intelligent fault diagnosis method based on multi-scale symbolic Lempel-Ziv (MSLZ) is proposed to identify the several faults of railway vehicle systems (RVS). The proposed MSLZ is essentially for the purpose of estimating the irregularity of a given time series. In the proposed MSLZ method, the symbolization and multi-scale techniques are combined with Lempel-Ziv (LZ) to enhance its feature extraction ability. First, the symbolization can facilitate the LZ to remove the noises and reserve the fault information. Second, multi-scale analysis can extend LZ to multiple time scales, which can further enhance the description ability of dynamic characteristics. Using numerical data and experimental signals collected from RVS, the performance of the MSLZ method is demonstrated to be sensitive to periodical impulses and robust to environmental noise. Moreover, it has been demonstrated that MSLZ has superiority in extracting fault information of RVS compared with LZ, symbolic Lempel-Ziv (SLZ), and multi-scale Lempel-Ziv (MLZ) methods.

The mass spectrometer filament current sensor reflects the health status of the filament by detecting real-time variation in filament current, providing an ideal means to monitor the safe use of the filament. However, the complex and changing environment inside and outside the sensor often affects the sensor, so it is important to perform fault diagnosis and calibration of the sensor. To address the problems of low fault identification accuracy and high calibration error of a single sensor due to insufficient fault samples, a CNN-LSTM-based fault diagnosis and adaptive multi-channel fusion calibration of filament current sensors for mass spectrometers is proposed. First, two valid sensor fault datasets are generated by numerical simulation of on-orbit historical data collected by a spacecraft through the CVAE model. Secondly, the fault identification of the dataset is performed by the CNN-LSTM model, and the designed weighted fusion loss function leads to the improved accuracy of the model. Finally, an adaptive multi-channel calibration model is constructed, and a method for adaptively selecting calibration channels based on the characteristics of the fault samples themselves is proposed, which significantly reduces the calibration error compared with a single calibration model. The experimental results show that the method can improve the fault identification rate and reduce the calibration fault error compared with other fault diagnosis and calibration methods. It provides an effective sensor fault solution for mass spectrometer filament current sensors.