Structural damage identification has been the focus of engineering fields, while the existing damage identification methods heavily depend on extracted “hand-crafted” features. Recently, due to the powerful feature learning capability of deep learning, it has been widely used in structural damage identification. However, those methods only consider the local dependence or temporal relation of data. Thus, in this paper, a structural damage identification method by combining the convolutional neural network (CNN) and gated recurrent unit (GRU) is proposed. The CNN model is used to extract the local dependence of data, and the GRU model is used to extract the temporal feature of data. These two extracted feature matrices are spliced horizontally to a fused eigenvector. The eigenvector is input to the final softmax classifier layer to identify the structural damage state. Experiments on a scale model of the three-span continuous rigid frame bridge shown that the CNN-GRU model performs significantly better than CNN, LSTM, and GRU models for structural damage identification.
Aiming at solving the problems of sensitivity to instrument shape and low accuracy of positioning by using the traditional feature matching points based methods, a novel positioning approach based on convolutional neural network is proposed in this work. The designed convolutional neural network consists of two convolutional layers, two pooling layers and two fully connected layers. The objective function adopts cross-entropy and the optimization method adopts Adam algorithm. We used 7000 images collected at different time and situations as test samples. We discussed the performance of the proposed algorithm under different amount of training data, different convolution kernel sizes and different number of convolution kernels in the experiments. Compared with the traditional feature matching point based method, the proposed method has higher recognition accuracy and lower false positioning rate.
Modern society is developing rapidly, new technological means are emerging, people’s mindset is changing day by day, and all aspects of social life have undergone great changes. In such an era, French teaching is facing new challenges, but it is also a new opportunity for teaching reform. The introduction of new deep learning model teaching tools in traditional teaching, the construction of online education, and the adoption of a combination of online and offline teaching models enable French teaching to adapt to the characteristics and needs of the new era and ensure the quality of teaching. The model uses deep learning as technical support and takes the prediction of students’ answers to a question as a judgment indicator. Through the generation of directed graphs in the model, the model structure is optimized to improve its application in predicting students’ answers compared to the classical model. It is prepared for the next step of building a system for personalized question recommendation using deep reinforcement learning. The model plays a role as a student simulator in the recommender system. So its prediction effect reflects the simulation effect on students. Finally, it is experimentally demonstrated that the proposed deep knowledge tracking model based on directed graphs has a significant improvement in prediction effectiveness compared to the traditional model.
scite is a Brooklyn-based organization that helps researchers better discover and understand research articles through Smart Citations–citations that display the context of the citation and describe whether the article provides supporting or contrasting evidence. scite is used by students and researchers from around the world and is funded in part by the National Science Foundation and the National Institute on Drug Abuse of the National Institutes of Health.