Under the circumstances of global warming and rapid urbanization, damage caused by urban inundation are becoming increasingly severe, attracting the attention of both researchers and governors. The accurate simulation of urban inundation is essential for the prevention of inundation hazards. In this study, a 1D pipe network and a 2D urban inundation coupling model constructed by InfoWorks ICM was used to simulate the inundation conditions in the typical urbanized area in the north of Lin’an. Two historical rainfall events in 2020 were utilized to verify the modeling results. The spatial–temporal variation and the causes of urban inundation under different designed rainfalls were studied. The results were as follows: (1) The constructed model had a good simulation accuracy, the Nash–Sutcliffe efficiency coefficient was higher than 0.82, R² was higher than 0.87, and the relative error was ±20%. (2) The simulation results of different designed rainfall scenarios indicated that the maximum inundation depth and inundation extent increased with the increase in the return period, rainfall peak position coefficient, and rainfall duration. According to the analysis results, the urban inundation in Lin’an is mainly affected by topography, drainage network (spatial distribution and pipe diameter), and rainfall patterns. The results are supposed to provide technical support and a decision-making reference for the urban management department of Lin’an to design inundation prevention measures.

Both constructivist learning and situation-cognitive learning believe that learning outcomes are significantly affected by the context or learning environments. However, since 2019, the world has been ravaged by COVID-19. Under the threat of the virus, many offline activities, such as some practical or engineering courses, have been subjected to certain restrictions. Virtual Reality (VR) is an emerging, promising, and rapidly developing technology that enables users to obtain a near-real immersion experience by combining technologies such as computer science, communication, vision, etc. In the context of COVID-19, the advantages of VR immersive experiences are highlighted. By constructing a virtual learning environment, VR technology can greatly compensate for the shortage of traditional teaching conditions and help learners to carry out cognitive learning better. However, currently, VR-enhanced cognitive learning is still in its infancy, along with numerous problems and limitations. Therefore, this paper first conducted an in-depth study of some related concepts, such as constructivist learning and situated cognition learning. Then it proposes a general VR-enhanced cognitive learning framework and designs the general steps for constructing learning situations with VR technology. Based on the proposed model and framework, it developed a campus knowledge-learning APP using VR tools. Through a case study, it verified the validity and performance of the model and strategies. Questionnaire survey and experimental results show that the new model achieves a good learning effect and improves the efficiency of learning by at least 20% compared to the traditional learning methods.

The intermittent nature of solar energy poses significant challenges to the integration of photovoltaic (PV) power generation into the electrical grid. Consequently, the precise forecasting of PV power output becomes essential for efficient real-time power system dispatch. To meet this demand, this paper proposes a deep learning model, the CA-Transformer, specifically designed for PV power output prediction. To overcome the shortcomings of traditional correlation coefficient methods in dealing with nonlinear relationships, this study utilizes the Copula function. This approach allows for a more flexible and accurate determination of correlations within time series data, enabling the selection of features that exhibit a high degree of correlation with PV power output. Given the unique data characteristics of PV power output, the proposed model employs a 1D-CNN model to identify local patterns and trends within the time series data. Simultaneously, it implements a cosine similarity attention mechanism to detect long-range dependencies within the time series. It then leverages a parallel structure of a 1D-CNN and a cosine similarity attention mechanism to capture patterns across varying time scales and integrate them. In order to show the effectiveness of the model proposed in this study, its prediction results were compared with those of other models (LSTM and Transformer). The experimental results demonstrate that our model outperforms in terms of PV power output prediction, thereby offering a robust tool for the intelligent management of PV power generation.