Wei Zhou

With the development of science and technology and the improvement of hardware computing power, the application of large models in the field of artificial intelligence (AI) has become a current research hotspot Among the focal points in the field of deep learning, AI for science is one of the highlighted areas, utilizing deep learning methods for pattern recognition, anomaly detection, predictive analysis, and more on a large scale of scientific data. In the realm of materials science, the structure of crystals is composed of edges and nodes, making it readily representable as a graph. In previous research, some typical models, such as the MEGNet model, utilized their graph neural network features to fit computational results based on density functional theory for predicting various material properties. Building on this concept, the authors propose a novel graph neural network (GNN) model, optimized with a Multi-Head Self-Attention (MHSA) mechanism, for predicting materials data with crystal structures. This model is named self-attention enhanced graph neural network. The model segments the input data into three parts: edges, nodes, and global features. The graph convolutional layer module is primarily used for aggregating node, edge, and global features, learning node representations, and capturing higher-order neighborhood information through multiple layers of GNN. The MHSA component allows nodes to learn global dependencies, providing different representation subspaces for the nodes. In comparison with other machine learning and deep learning models, the results indicate an improvement in the predictive accuracy of this model. A new graph neural network (GNN) model called Self-Attention Enhanced Graph Neural Network (SA-GNN) is proposed for predicting the properties of materials with crystal structures. This model incorporates multi-head self-attention to allow nodes to learn global dependencies and generate different representational subspaces. Compared to other machine learning and deep learning models, the results show improved predictive accuracy, demonstrating the potential of graph networks combined with self-attention for modeling crystal material data.

The unknown fault diagnosis technology in industrial systems implies significant engineering application value and opportunities. The difficulty stems from the fact that the unknown fault samples frequently originate from the diagnostic model’s unknow distribution, leading to an out-of-distribution (OOD) problem. An incorrect diagnosis in the diagnostic model might readily arise from this. To deal with this problem, this article proposes a novel trustworthy fault diagnosis with OOD detection which can be applied on industrial systems and equipment. First, deep base learners (DBLs) with different activation functions are designed to construct the deep ensemble model. After that, use in-distribution (ID) inputs to train the initial deep ensemble model. Then, with the proposed max consistency and min similarity guided criterion, the DBLs of the initial ensemble model are chosen to reconstruct the ensemble model. Finally, the diagnostic results’ uncertainty of the reconstruct ensemble model is estimated to accurately determine the type of the sample to be diagnosed. To verify the effectiveness of the proposed method, two gearbox data sets were used to test the proposed method and the max consistency and min similarity guided criterion. The experimental results demonstrate that the proposed approach can accurately identify unknown fault samples in the gearbox.