首页|基于遥感初判的建筑物震害Fisher判别法研究

基于遥感初判的建筑物震害Fisher判别法研究

扫码查看
破坏性地震发生后,迅速准确地预测建筑物破坏程度,对快速科学地开展地震应急指挥、救援力量部署等工作具有重大意义.针对现有建筑震害预测模型存在的评估结果粗糙、数据获取难度较高、计算工作量大、模型构建难度大、普适性不强等问题,提出一种基于遥感初判的建筑物震害Fisher判别法.首先,从地震强度和建筑物抗震能力两方面选取震级、震中距、场地条件及建筑物抗震能力4种震害作为判别因子;然后,基于判别分析理论,构建建筑物震害Fisher判别模型;最后,以四川泸定县6.0级地震为例,对文章提出方法进行验证.实验结果表明:该方法的预测结果与实际震害基本一致,准确率高达80%以上,证明该方法具有较高的可靠性,能较准确地对建筑物震害程度进行科学预测.
A Fisher discriminant method for seismic damage of buildings based on preliminary judgment from remote sensing
After a destructive earthquake,quickly and accurately predicting the damage degree of buildings is of great importance to quickly and scientifically carry out earthquake emergency com-mand and rescue force deployment.Existing building seismic damage prediction models yield im-precise evaluation results,struggle with data acquisition,and require a large amount of compu-ting power;therefore,they are difficult to construct and lack universality.To address these is-sues,this paper proposes a Fisher discriminant method for seismic damage of buildings based on preliminary judgment from remote sensing.First,four seismic damage factors,including magni-tude,epicentral distance,site condition,and seismic resistance of the building,were selected as the discriminant factors.Then,based on the discriminant analysis theory,the Fisher discriminant model of building damage was constructed.Finally,the method proposed in this paper was veri-fied by taking the Luxian M6.0 earthquake as an example.The experimental results show that the prediction results of the proposed method closely align with the actual earthquake damage,with an accuracy rate of 80%,which proves that this method is highly accurate,reliable,and can ac-curately predict the earthquake damage degree of buildings.

preliminary judgment from remote sensingbuildingsseismic damage predictionFisher discriminant methodLuxian earthquake

赵真、郭红梅、张莹、尹文刚、鲁长江、范开红、张翼

展开 >

四川省地震局,四川成都 610041

武警警官学院,四川成都 610041

遥感初判 建筑物 震害预测 Fisher判别法 泸定县地震

2024

地震工程学报
中国地震局兰州地震研究所,中国地震学会,清华大学,中国土木工程学会

地震工程学报

CSTPCD北大核心
影响因子:1.191
ISSN:1000-0844
年,卷(期):2024.46(6)