摘要
为了更加全面、准确地对山火风险进行评估,提高架空输电线路周边的火灾管理水平,从地形、人为因素、遥感和气象4个方面选取高程、坡度、坡向、植被覆盖类型、距最近道路距离、距最近水体距离、距最近居民点距离、地表温度、可燃物含水率、归一化植被指数、降水量、平均气温、最低气温、最高气温、平均相对湿度、最大阵风风速、最大阵风风向这17个相关评价因子对山火进行评估风险。然后应用相关性分析,剔除相关系数大于0.6的平均气温和最低气温2个因子。最后以云南省为研究区域,基于Logistic回归模型,协同所选15个因子对云南省山火风险进行评估。实验结果表明,所提出的协同多因子山火风险评估方法优于基于其他相关因子的山火评估方法。
In order to comprehensively and accurately evaluate the wildfire risks and improve the fire management level around the overhead transmission lines,seventeen assessment factors,including the elevation,the slope,the aspect,the land cover type,the distance to the nearest path,the distance to the nearest water body,the distance to the nearest residential area,the land surface temperature,the fuel moisture content,the normalized differential vegetation index,the precipitation,the average temperature,the minimum temperature,the maximum temperature,the average relative humidity,the maximum gust wind speed,the maximum gust wind aspect,are selected in terms of topography,human factors,remote sensing and meteorology to evaluate the wildfires.Through correlation analysis,two redundant factors of the average temperature and the minimum temperature with the coefficient value being greater than 0.6 are eliminated.Finally,Yunnan province is taken as the research area,the wildfire risks are evaluated based on the Logistic regression model in consideration of the above 15 factors.The experimental results show that the cooperative multi-factor wildfire risk assessment method proposed is superior to the assessment method based on other related factors.
作者
文刚
高孟平
高振宇
周仿荣
潘浩
李志栋
张涛
罗继强
刘星星
WEN Gang;GAO Mengping;GAO Zhenyu;ZHOU Fangrong;PAN Hao;LI Zhidong;ZHANG Tao;LUO Jiqiang;LIU Xingxing(Joint Laboratory of Power Remote Sensing Technology(Electric Power Research Institute,Yunnan Power Grid Co.,Ltd.),Kunming,Yunnan 650217,China;Yunnan Power Grid Co.,Ltd.,Kunming,Yunnan 650200,China;Beijing Institute of Spacecraft System Engineering,Beijing 100094,China;Guangdong University of Technology,Guangzhou,Guangdong 510006,China)
出处
《广东电力》
2022年第4期96-102,共7页
Guangdong Electric Power
基金
广州市科技计划项目(201902020012)。
关键词
山火
遥感
气象
评估
LOGISTIC回归
wildfire
remote sensing
meteorology
evaluation
Logistic regression
