2010年中国大陆100m分辨率用电量栅格数据集

本数据集为2010年中国大陆100m分辨率用电量栅格数据集，共包含两份数据，分别为工业用电量100m分辨率网格数据集和非工业用电量100m分辨率网格数据集，数据单元单位均为万千瓦时/公顷。本研究利用随机森林回归算法，对工业用电量和非工业用电量分别进行建模和预测。其中由于用电量数据缺测较多，鉴于GDP和用电量的高相关性，利用工业GDP和非工业GDP数据来预测地级市的工业和非工业用电量。将工业兴趣点与夜间灯光、NDVI、海拔以及路网等数据作为辅助数据，基于随机森林模型，构建工业用电量模型；同样，基于非工业兴趣点和多源遥感数据作为辅助数据，构建非工业用电量模型；最后利用分区密度制图方法，对工业用电量和非工业用电量进行空间分配。本研究得到的高精度网格化用电量数据，对于正确认识中国大陆用电量分布，合理配置和有效利用电力资源具有重要意义。

This dataset is a 100-meter-resolution gridded electricity consumption dataset for mainland China in 2010, which includes two subsets: the 100-meter-resolution grid dataset of industrial electricity consumption and the 100-meter-resolution grid dataset of non-industrial electricity consumption. The unit of each grid cell is 10,000 kilowatt-hours per hectare.
This study adopted the random forest regression algorithm to separately model and predict industrial and non-industrial electricity consumption. Given the high missing rate of original electricity consumption data and the strong correlation between GDP and electricity consumption, industrial GDP and non-industrial GDP data were used to predict the industrial and non-industrial electricity consumption at the prefecture-level city scale. For the construction of the industrial electricity consumption model, industrial points of interest (POIs), nighttime light data, NDVI, elevation, road network and other auxiliary data were incorporated into the random forest framework. Similarly, the non-industrial electricity consumption model was built using non-industrial POIs and multi-source remote sensing data as auxiliary variables. Finally, spatial allocation of industrial and non-industrial electricity consumption was conducted via the zone density mapping method.
The high-precision gridded electricity consumption data derived from this study is of great significance for accurately understanding the spatial distribution of electricity consumption in mainland China, as well as rationally allocating and effectively utilizing power resources.

搜集汇总

背景与挑战

背景概述

该数据集是2010年中国大陆100米分辨率的用电量栅格数据，包含工业用电和非工业用电两个子集，数据单位为每公顷万千瓦时。它基于随机森林回归算法，结合GDP和多源遥感辅助数据（如POI和夜间灯光）建模生成，旨在提供高精度的用电空间分布，对理解中国电力资源分配和利用具有重要意义。