📄 中文摘要
在能源系统分析中，耦合具有不匹配空间分辨率的模型是一项重大挑战。常见的解决方案是为高分辨率地理单元分配权重以进行聚合，但传统模型仅使用单一地理属性，限制了其有效性。研究提出了一种创新方法，采用自监督的异构图神经网络来解决这一问题。该方法将高分辨率地理单元建模为图节点，整合多种地理特征，为每个网格点生成具有物理意义的权重。这些权重增强了传统的Voronoi基分配方法，使其不仅限于地理接近性，还能纳入重要的地理信息。
利用图神经网络改善能源系统耦合的空间分配
📄 English Summary
Improving Spatial Allocation for Energy System Coupling with Graph Neural Networks
In energy system analysis, coupling models with mismatched spatial resolutions poses a significant challenge. A common approach involves assigning weights to high-resolution geographic units for aggregation; however, traditional models are constrained by utilizing only a single geospatial attribute. An innovative method is proposed that employs a self-supervised Heterogeneous Graph Neural Network to tackle this issue. This method models high-resolution geographic units as graph nodes, integrating various geographical features to generate physically meaningful weights for each grid point. These weights enhance the conventional Voronoi-based allocation method, allowing it to transcend mere geographic proximity by incorporating essential geographic information.