遥感数据驱动的电力污秽等级XGBOOST预测模型  

XGBOOST Prediction Model for Power Fouling Level Based on Remote Sensing Data-driven

在线阅读下载全文

作  者:周仿荣 文刚 马仪 张辉 朱龙昌 杨可意 韩舸 ZHOU Fangrong;WEN Gang;MA Yi;ZHANG Hui;ZHU Longchang;YANG Keyi;HAN Ge(Joint Laboratory of Power Remote Sensing Technology,Electric Power Research Institute,Yunnan Power Grid Co.,Ltd.,China Southern Power Grid,Kunming 650217,China;School of Remote Sensing and Information Engineering,Wuhan University,Wuhan 430079,China)

机构地区:[1]电力遥感技术联合实验室(南方电网公司云南电网电力科学研究院),云南昆明650217 [2]武汉大学遥感信息工程学院,湖北武汉430079

出  处:《地理空间信息》2023年第12期21-24,共4页Geospatial Information

基  金:云南省重大科技专项资助项目(202202AD080010)。

摘  要:输电线路绝缘子积污是电网安全运行的重要威胁之一。传统的电力污秽等级评估依靠人工作业,存在效率低、准确性不足等问题。遥感和人工智能技术的发展为改进电力污秽等级预测模型提供了新的契机。本研究搜集了夜间灯光遥感数据、大气环境遥感监测数据、NDVI和污染源排放清单等多源异构数据,开发了一种基于XGBOOST算法的电力污秽等级预测模型,用于实现大范围、细粒度的电力系统污区图绘制。实验结果表明,该方法在测试集上取得了87%的总体精度,对于电力系统关注的重污染区域预测精度在82%以上。模型所需的数据源可以通过公开渠道免费获得,因此该方法将为今后电力系统低成本、高效、准确的绘制电力污区图提供重要支撑。Transmission line insulator fouling accumulation is one of the important threats to the safe operation of power grids.Traditional power fouling level assessment relies on manual work,which has problems such as low efficiency and accuracy.The development of remote sensing and artificial intelligence technology provides new opportunities to improve the prediction model of power fouling level.During the research,we collected multi-source heterogeneous data such as remote sensing data of nighttime lights,remote sensing monitoring data of atmospheric envi-ronment,NDVI and emission inventory of pollution sources,and developed a prediction model of power fouling level based on XGBOOST algo-rithm for realizing large-scale and fine-grained fouling area mapping of power system.The experimental results show that the method achieves an overall accuracy of 87%on the test set and a prediction accuracy of over 82%for the heavily polluted areas of power system concern.The da-ta sources required by the proposed model in this paper are freely available through public channels,so the method will provide important sup-port for low-cost,efficient and accurate mapping of power pollution areas in power systems in the future.

关 键 词:电力污秽 夜光遥感 ESDD XGBOOST 

分 类 号:P237[天文地球—摄影测量与遥感]

 

参考文献:

正在载入数据...

 

二级参考文献:

正在载入数据...

 

耦合文献:

正在载入数据...

 

引证文献:

正在载入数据...

 

二级引证文献:

正在载入数据...

 

同被引文献:

正在载入数据...

 

相关期刊文献:

正在载入数据...

相关的主题
相关的作者对象
相关的机构对象