2020 年中国地面光伏电站 10 米全国比例尺地图

中国农业大学与中国科学院地理科学与资源研究所合作，针对国内缺乏高分辨率、开源的全国地面光伏电站分布数据现状，开发了一套全国性的制图方法。该方法采用分区建模、随机森林分类技术，并结合主动学习策略进行优化，利用欧空局哨兵 2 号卫星的多光谱影像，成功发布了 2020 年全国 10 米分辨率地面光伏电站分类产品。该成果不仅精确揭示了我国光伏电站的空间分布特征，而且以 89% 的分类精度，为能源规划、土地利用、遥感监测及环境研究等领域提供了宝贵的数据资源。此举填补了国内该领域数据的空白，对相关研究具有重要意义。

Cooperated by China Agricultural University and the Institute of Geographic Sciences and Natural Resources Research, Chinese Academy of Sciences, a research team developed a national-scale mapping methodology to address the current lack of high-resolution, open-source nationwide ground-mounted photovoltaic (PV) power station distribution datasets in China. This approach adopts zonal modeling and random forest classification techniques, and is optimized with an active learning strategy. Using multispectral imagery from the European Space Agency's Sentinel-2 satellites, the team successfully released the 2020 national 10-meter resolution ground-mounted PV power station classification product. This product accurately reveals the spatial distribution characteristics of PV power stations in China, achieving a classification accuracy of 89%, and provides valuable data resources for fields including energy planning, land use, remote sensing monitoring and environmental research. This work fills the domestic data gap in this field, and holds important significance for relevant research.

搜集汇总

背景与挑战

背景概述

该数据集是2020年中国地面光伏电站的10米全国比例尺地图，由中国农业大学与中国科学院地理科学与资源研究所合作开发，采用随机森林分类和主动学习技术，基于哨兵2号卫星影像生成，分类精度达89%。它提供了高分辨率的光伏电站空间分布数据，适用于能源规划、土地利用和遥感监测等研究，填补了国内该领域的数据空白。