In this paper, a fault diagnosis algorithm named improved one-dimensional deep residual shrinkage network with a wide convolutional layer (1D-WIDRSN) is proposed for quadrotor propellers with minor damage, which can effectively identify the fault classes of quadrotor under interference information, and without additional denoising procedures. In a word, that fault diagnosis algorithm can locate and diagnose the early minor faults of the quadrotor based on the flight data, so that the quadrotor can be repaired before serious faults occur, so as to prolong the service life of quadrotor. First, the sliding window method is used to expand the number of samples. Then, a novel progressive semi-soft threshold is proposed to replace the soft threshold in the deep residual shrinkage network (DRSN), so the noise of signal features can be eliminated more effectively. Finally, based on the deep residual shrinkage network, the wide convolution layer and DroupBlock method are introduced to further enhance the anti-noise and over-fitting ability of the model, thus the model can effectively extract fault features and classify faults. Experimental results show that 1D-WIDRSN applied to the minimal fault diagnosis model of quadrotor propellers can accurately identify the fault category in the interference information, and the diagnosis accuracy is over 98%.

This paper introduces a new intelligent fault diagnosis method based on stack pruning sparse denoising autoencoder and convolutional neural network (sPSDAE-CNN). This method processes the original input data by using a stack denoising autoencoder. Different from the traditional autoencoder, stack pruning sparse denoising autoencoder includes a fully connected autoencoding network, the features extracted from the front layer of the network are used for the operation of the subsequent layer, which means that some new connections will appear between the front and rear layers of the network, reduce the loss of information, and obtain more effective features. Firstly, a one-dimensional sliding window is introduced for data enhancement. In addition, transforming one-dimensional time-domain data into the two-dimensional gray image can further improve the deep learning (DL) ability of models. At the same time, pruning operation is introduced to improve the training efficiency and accuracy of the network. The convolutional neural network model with sPSDAE has a faster training speed, strong adaptability to noise interference signals, and can also suppress the over-fitting problem of the convolutional neural network to a certain extent. Actual experiments show that for the fault of unmanned aerial vehicle (UAV) blade damage, the sPSDAE-CNN model we use has better stability and reliable prediction accuracy than traditional convolutional neural networks. At the same time, For noise signals, better results can be obtained. The experimental results show that the sPSDAE-CNN model still has a good diagnostic accuracy rate in a high-noise environment. In the case of a signal-to-noise ratio of −4, it still has an accuracy rate of 90%.

In this article, a fault-tolerant control method based on augmented improved extended state observer and non-singular high-order fast terminal sliding mode is proposed for a class of second-order systems with actuator faults. First, the initial peak of the traditional extended state observer is avoided by improving the structure of the observer. Second, the total disturbance and its change trend are observed simultaneously, so as to better realize the compensation of total disturbance. The convergence of the observer is proved theoretically. In addition, by designing non-singular high-order fast terminal sliding mode surface, the sliding mode variable converges rapidly during the whole process to improve the algorithm’s rapidity. Finally, the chattering of control signal caused by sliding mode control is greatly reduced using high-order sliding mode technology, and the stability of the whole closed-loop system is proved by Lyapunov criterion. The comparative experimental results on the fault-tolerant control platform of the quadrotor unmanned aerial vehicle demonstrate the effectiveness and superiority of the proposed observer and controller.

As the core component of rotating machinery, the fault diagnosis of rolling bearing has important engineering practical significance. Most of the current intelligent fault diagnosis methods are based on the premise that the training data and test data have similar probability distributions. However, in practical scenarios, there will inevitably be discrepancies in the distribution of vibration signals due to internal and external factors such as changes in working conditions, which will significantly affect the diagnostic performance of the intelligent diagnostic model. Aiming at problems that the vibration signal characteristic distribution of rolling bearings is inconsistent under different working conditions and the labels of the samples to be diagnosed are difficult to obtain, a new domain-adaptive fault diagnosis method is proposed in this paper. Firstly, the multi-scale feature extraction module is used to extract the features of the input signals, and the residual network structure is used to avoid the degradation of the model performance. Then, the APReLU activation function is used to make the vibration signals perform different nonlinear transformations according to their own characteristics through adaptive learning. Finally, the Joint Maximum Mean Discrepancy (JMMD) is used to reduce the displacement of both conditional and edge distributions between different domains. Therefore, this method can extract domain-invariant feature information and align the source and target domains, which can be used for cross-domain intelligent fault diagnosis. Six transfer fault diagnosis tasks based on the rolling bearing experimental platform are designed to evaluate the performance and effectiveness of the proposed method. At the same time, four popular methods are selected for comprehensive analysis and comparison. The results show that the method has good robustness and superiority under various diagnostic tasks.