Enhancing the accuracy of short-term wind power forecasting can be effectively achieved by considering the spatial–temporal correlation among neighboring wind turbines. In this study, we propose a short-term wind power forecasting model based on 3D CNN-GRU. First, the wind power data and meteorological data of 24 surrounding turbines around the target turbine are reconstructed into a three-dimensional matrix and inputted into the 3D CNN and GRU encoders to extract their spatial–temporal features. Then, the power predictions for different forecasting horizons are outputted through the GRU decoder and fully connected layers. Finally, experimental results on the SDWPT datasets show that our proposed model significantly improves the prediction accuracy compared to BPNN, GRU, and 1D CNN-GRU models. The results show that the 3D CNN-GRU model performs optimally. For a forecasting horizon of 10 min, the average reductions in RMSE and MAE on the validation set are about 10% and 11%, respectively, with an average improvement of about 1% in R. For a forecasting horizon of 120 min, the average reductions in RMSE and MAE on the validation set are about 6% and 8%, respectively, with an average improvement of about 14% in R.

The first law of geography is one of the most important concepts in geographical analysis, revealing the significant role of spatial proximity. At present, some current international relation studies or geographic network analysis studies tend to build corresponding network models according to different themes, but the most basic level of geographic neighborhoods is intentionally or unintentionally neglected in those processes. Based on the adjacency relationship between the terrestrial countries in the world, the model of the terrestrial adjacency network (TAN) is constructed. The model includes almost all land-based countries and is divided into three main regions, respectively, Eurasia, Africa, and America. On the mathematical model of these regions, we analyze the geospatial structure and network evolution of the adjacent networks utilizing statistical methods and network analysis methods. This study helps to map and understand the geographical attributes and characteristics of countries from the perspective of holistic structure, aiming to provide a quantitative reference for subsequent research on international relations and geographic computing. Moreover, despite some limitations, TAN represents a new advance in geographical network analysis that can be further applied by overlaying more attribute data.

The uncertainty of COVID-19 and the spatial inequality of anti-pandemic materials have made international aid an important means for many countries to cope with this global public health crisis. It is of far-reaching significance to analyze the spatiotemporal characteristics and influencing factors of international aid events for the global joint fight against COVID-19 and the sustainability of global public health business. The data on aid events from 23 January 2020 to 31 October 2021, were from the GDELT database. China, the United States, the United Kingdom, and Canada were selected as the study objects because they provided more aid. Their spatiotemporal characteristics of main aid flows, the response characteristics of the aid requests, and the characteristics of verbal aid to cash in were studied using spatial statistical analysis methods. The influencing factors of aid allocation also were studied by regression analysis. The results found that: the international aid flow of each country was consistent in spatial distribution, mainly to countries with severe pandemics and neighboring countries. However, there were differences in the recipients. China mainly aided developing countries, while the United States, the United Kingdom, and Canada mainly aided developed countries. Relatively speaking, China was more responsive to aid requests and more aggressive in cashing in on verbal aid. The countries considered the impact of their economic interests when they planned to aid. At the same time, there were obvious “bandwagon effect” and “small country tendency” on the aid events.

To overcome the limitations of long-term prediction of PM_2.5 concentration, a multi-factor information flow causality analysis method is used to screen suitable meteorological and air pollutant-related factors and concatenate them with a PM_2.5 sequence as the dataset. A modal decomposition algorithm is used as a module to be integrated into the autoformer (transformer improved with autocorrelation mechanism) model to improve it, and the modal autoformer (empirical modal decomposition combined with autoformer) is proposed. The constructed model decomposes the sequence into several components by using the modal decomposition module and uses the self-correlation mechanism and decomposition structure to decompose and extract features of different components at the time-feature level. Based on the matching method, the model is adjusted for different component features to improve the long-term prediction effect. The model is applied to three cities in Henan Province, Zhengzhou, Luoyang, and Zhumadian, as examples for experiments, and gated neural unit (GRU), informer, autoformer, and modal GRU (empirical modal decomposition combined with GRU model) are constructed for comparative verification. The results show that the modal autoformer can better cope with the complex characteristics of long-term prediction of the PM_2.5 time series, has strong spatial adaptability and that its various indicators are optimal for the three cities, with R² values being all above 0.96, where the highest is 0.987 in Zhengzhou; MAPE (Mean absolute percentage error) values all being less than 10, where the best is 7.602 in Zhumadian; and MAE (Mean absolute error) values all being less than 4. The prediction effect is stable enough, showing its feasibility and adaptability in long-term prediction.

The study of cross-border transport connectivity is significant for the development of regional integration and insight into global patterns. Comprehensive connectivity evaluations are lacking and insufficient attention has been paid to Latin American connectivity, so it is of great practical importance to comprehensively and rationally evaluate Latin American connectivity. In this article, based on the four modes of transport, namely, sea, road, air and railroad, and using the actual trade volume as a comparison, a connectivity evaluation index system with considerable reliability and generalization ability was constructed using the expert scoring method, QAP correlation analysis, QAP regression, and statistics, and the connectivity calculations of Latin America were obtained. Analyzing the connectivity structure of Latin America, it was found that cross-border passenger and cargo transport in the region was dominated by sea transport and supplemented by road and air transport, with railroads used the least. The overall connectivity of Latin America was low, and the overall development was unbalanced, with a strong law of spatial differentiation, which was mainly manifested in the strongest connectivity of the integrated coastal countries, followed by the island countries, and the lowest connectivity of the landlocked countries. Different countries assumed different roles in regional connectivity, which could be categorized into global hub type, local hub type and non-hub type based on the calculations. There was a spatial pattern of decreasing connectivity with distance in typical countries, but the rate of decline was closely related to their geographic location and the role they played in the connectivity network. This study can provide reference and inspiration for regional connectivity evaluation, improvement, and sustainable development.