Wind power is of strategic importance for reducing carbon dioxide emissions, minimizing environmental pollution, and enhancing the sustainability of energy supply. Health monitoring of wind turbines is a crucial technology to ensure the quality of grid-connected power. Insufficient labeled data and class imbalance problems are two critical issues for intelligent fault diagnosis of wind turbines. In this article, an intelligent fault diagnosis method based on stacked capsule autoencoders is proposed to address the issues of inadequate labeled data and class imbalance. A prior knowledge-based convolution layer is applied to optimize the initialization of capsules, making it more conducive to learning spectral information. The pose representations of parts and objects can be improved, and a method for embedding spectral templates is proposed. The stacked capsule autoencoder in this study can learn partial templates unsupervised through likelihood estimation and establish the mapping between capsules and fault types. The experimental results, obtained from the CWRU dataset and a private dataset from a wind turbine drive-train simulation platform, demonstrate that the proposed method is robust to imbalanced and small-sized datasets. It can perform stable and effective unsupervised training by utilizing a sufficient amount of normal class data to expedite learning convergence.

Automated retinal vessel segmentation is crucial for computer-aided clinical diagnosis and retinopathy screening. However, deep learning faces challenges in extracting complex intertwined structures and subtle small vessels from densely vascularized regions. To address these issues, we propose a novel segmentation model, called Geometry-Knowledge Embedded TransUNet (GKE-TUNet), which incorporates explicit embedding of topological features of retinal vessel anatomy. In the proposed GKE-TUNet model, a skeleton extraction network is pre-trained to extract the anatomical topology of retinal vessels from refined segmentation labels. During vessel segmentation, the dense skeleton graph is sampled as a graph of key-points and connections and is incorporated into the skip connection layer of TransUNet. The graph vertices are used as node features and correspond to positions in the low-level feature maps. The graph attention network (GAT) is used as the graph convolution backbone network to capture the shape semantics of vessels and the interaction of key locations along the topological direction. Finally, the node features obtained by graph convolution are read out as a sparse feature map based on their corresponding spatial coordinates. To address the problem of sparse feature maps, we employ convolution operators to fuse sparse feature maps with low-level dense feature maps. This fusion is weighted and connected to deep feature maps. Experimental results on the DRIVE, CHASE-DB1, and STARE datasets demonstrate the competitiveness of our proposed method compared to existing ones.

This articleproposes an intelligent algorithm for wind turbine faults identification based on the image model of the dynamic process and the deep convolutional neural network. First, feature engineering is designed to generate the image model of a dynamic process. We performed variable refinement and data normalization preprocessing on the dataset. Then the time series data are reconstructed in the Gram angle field to form an image model. Next, an identification algorithm adapted to the image model, called multistream self-fusion net (MuSnet), is proposed. Inside the MuSnet, we use the multistream self-fusion module to replace the original convolution operation in the low-level convolution. This allows the features of heterogeneous information in the image model to be better extracted. Multiple evaluation metrics in the experiment show that the proposed method has advantages of high recognition accuracy, ability to train with fewer samples, speeding up the convergence of learning, and high robustness for wind turbine fault identification.

The data loss and corresponding false fault features of single-phase or multiphase detection signals caused by sensors is a relatively troublesome problem, which can increase the difficulty to the fault diagnosis of switch fault and open-phase fault of inverter, and even can lead to false alarm. For these problems, a complementary virtual mirror fault diagnosis method for microgrid inverter is proposed to improve them. First, the virtual mirrors are constructed, which contain fault information in different angle. Second, a cross comparison processing method is used to gain the cross variables and corresponding mirror cross variables by the obtained virtual mirrors and detected signals, which can extract and normalize fault components from different angle in the same way, respectively. Thus, it can reduce the impact of data loss. Third, the self-correction fault degree (scf) and averages fault degree (af) are calculated. Specially, scf can further reduce the influence of data loss and false fault features through the complementarity of cross variables and mirror cross variables. For af, it is a steady-state expression of fault degree, which can reduce the fluctuation of fault degree variable and the possibility of false alarm. Then, fault detection is implemented through the joint decision of scf and af. Finally, the fault is located by the fault detection results and the extracted mirror complementary location information. Compared with the existing fault diagnosis methods, it not only can diagnose switch fault of inverter, but also can have certain robustness for single-phase or multiphase data loss. Meanwhile, it also has relatively fewer parameters and the lower difficulty of debugging, which are conducive to practical application. The effectiveness of the proposed method is validated by the experiment results.

The optimal control of fuzzy systems with constraints is still an open problem. Our focus concerns the optimal control problem of fuzzy systems derived from receding horizon control (RHC) schemes. We consider methods to numerically compute the value function for general fuzzy systems. The numerical method that is developed using the finite difference with sigmoidal transformation is a stable and convergent algorithm for the Hamilton-Jacobi-Bellman (HJB) equation. An optimization procedure is developed to increase the calculation accuracy with less computation time. A parallel-processing method is employed in the optimization procedure. The optimization results are applied to the controller design of general fuzzy dynamic systems. Employing the principle of conventional RHC schemes, RHC-form controllers are designed for some classes of fuzzy dynamic systems. The basic ideas are as follows. First, the value function is calculated by numerical methods. Then, the value function is used as controller-design parameters to redesign RHC controllers for fuzzy systems, which is motivated by the inverse Lyapunov function design method. It is proven that the closed-loop system is asymptotically stable. An engineering implementation of the controller redesign scheme is discussed. Meanwhile, the parallel-processing framework that can improve the closed-loop performance is also introduced.

In this correspondence, we show that the relationship between the finite optimal trajectory and the infinite optimal trajectory in Lemma 3 of the aforementioned paper cannot act as a theory basis for the further implementation of other theorems and the dynamic decomposition algorithm.