The rolling bearing is the key component of rotating machinery, and it is also a failure–prone component. The intelligent fault diagnosis method has been widely used to accurately diagnose bearing faults. However, in engineering practice, it is difficult to obtain sufficient sample data to train the intelligent diagnosis model. Therefore, in this article, a fusion diagnosis model CGAN-2-D-CNN that combines a conditional generative adversarial network (CGAN) and a two-dimensional convolutional neural network (2-D-CNN) is proposed for bearing fault diagnosis with small samples. Considering the problem of insufficient sample data, CGAN is used to learn the data distribution of real samples to generate new samples with similar data distribution by the confrontation training of the generator and discriminator. Then, 2-D preprocessing is conducted to convert the generated 1-D data into 2-D gray images. Finally, these images are input into the 2-D-CNN to extract the features and classify the bearing fault types. Two experimental cases are implemented to validate the effectiveness and feasibility of the proposed CGAN-2-D-CNN. The experimental results illustrate that the diagnosis accuracy of the proposed method used on the small sample data is close to that of the 2-D-CNN directly used on the enough original sample data whose size is equal to the expanded sample size. In addition, compared with the 1-D-CNN, support vector machine (SVM), and long short-term memory (LSTM) models, the 2-D-CNN model after 2-D preprocessing has the higher fault classification accuracy.

The intelligent diagnosis model based on deep learning method can effectively and accurately diagnose the health state of bearings widely used in various mechanical equipment. However, in engineering practice, it is difficult to obtain sufficient labeled fault data, which would reduce the diagnosis performance. In order to solve the above problem, this paper proposes a self-attention metric learning based on multi-scale feature fusion to classify the bearing fault with few shot. The proposed few-shot intelligent diagnosis model mainly contains the feature extraction module and metric learning module. First, the vibration data need to be combined in pairs to form sample pairs, which are input into the feature extraction module based on multi-scale feature fusion. This input approach can increase the learning tasks so as to alleviate the problem of insufficient training samples. Then, the extracted features are concatenated and input into the metric learning module whose learning ability is improved by adding the self-attention network, and the input features are eventually converted into the similarity. Finally, the fault type can be identified according to the maximum similarity value between the test sample and samples with different labels. Three experiment cases are conducted to validate the performance of proposed intelligent diagnosis model. The experimental results show that the proposed model can accurately classify the bearing fault with few-shot case. Additionally, the comparison experiments with the traditional models have also validated the advantage of the proposed intelligent diagnosis model for few-shot case.

Predicting the remaining useful life (RUL) of rolling bearings is a critical technology for ensuring safe operation and reducing the maintenance costs of rotating mechanical equipment. The accuracy of RUL prediction is highly dependent on the quality of degradation features screened from the original statistical features. However, there is redundancy between degradation features after screening, and the screening rule is highly subjective. Additionally, some useful information of the unselected features is lost. To solve the above problems, this paper proposes a feature fusion-based method for bearing RUL prediction. First, the self-organizing map (SOM) method is used to cluster the statistical features, which are then subjected to dimensionality reduction by the principal component analysis (PCA) method to obtain a fusion degradation feature set. Second, a BiLSTM network combined with the self-attention (SA) mechanism (BiLSTM-SA model) is established to predict bearing RUL with the fusion feature set. The SA mechanism is introduced to assign different weights to different fusion features to adaptively obtain the optimal feature combination. Finally, the effectiveness of the proposed feature fusion-based prediction method is verified on the FEMTO bearing dataset and IMS bearing dataset. The experimental results show that the proposed method can accurately predict bearing RUL, and its performance is superior to that of some state-of-the-art methods.

Space arm system is a typical multibody mechanical system requiring high operation stability and precision connected by kinematic joints. As joint clearance is inevitable during the manufacturing and mounting process, it turns into a critical problem affecting the dynamic characteristics and kinematic accuracy of the mechanism. A space arm system deploying stably and accurately is the prerequisite to guarantee the final satellite pointing accuracy once it is into its working state. Thus, the space arm system needs to have good dynamic characteristics and stability during its deploying process and to have accurate and reliable positioning characteristics during its locking process. For the deploying process, the mathematical model of a multibody system with clearance joint has been established as a foundation to study the influence of clearance properties, including clearance size, clearance joint locations, numbers of clearance joints and lubrication on the dynamic performance and motion stability of a two-degrees-of-freedom space arm system. The mathematical model of a locking mechanism has been established to study the clearance effects on the pointing accuracy of the space arm system during its locking process. The simulation results show that, the clearance has different effects on the dynamic characteristics of the space arm system during its different working processes.

The time history of the impact force, especially for the peak force, is vital to monitor the performance of mechanical products over the life-time. However, considering the limitation of sensing technology and inaccessibility of installing, it is always difficult or even impossible to measure the force directly in engineering practice. Therefore, a regularized cubic B-spline collocation (RCBSC) method combined with the modified L-curve criterion (RCBSC-ML) is presented to identify the impact force by easily measurable dynamic response. Because the profile of cubic B-spline is close to that of impact force, a linear combination of cubic B-spline functions is used to approximate the unknown impact force. Then the force identification equation is reformed, wherein the unknown variable changes from impact force vector to weight coefficient vector. Furthermore, the modified L-curve criterion is developed to select the optimal number of collocation points which is the regularization parameter of RCBSC method. Rich simulation studies on a five degree-of-freedoms structure and experimental studies on a simplified artillery test-bed are performed to validate the performance of RCBSC-ML method. The results illustrate that RCBSC-ML method can be used to accurately identify the unknown impact force, and compared with modified generalized cross validation criterion, the presented modified L-curve criterion is more suitable to determine the optimal regularization parameter of RCBSC method. Furthermore, compared with classical Tikhonov regularization method, RCBSC-ML method is more efficient and accurate in impact force identification.