基于网络灾情数据融合的震后建筑物倒塌快速评估  

Rapid Assessment of Post-earthquake Building Collapse Using Web-based Disaster Data Fusion

作　　者：闫泽正 赵晗萍 梁瑾璠 朱雷 刘宇翔 YAN Zezheng;ZHAO Hanping;LIANG Jinfan;ZHU Lei;LIU Yuxiang(Aerospace Science&Industry Haiying Group Co.,Ltd.,Beijing 100070,China;Key Laboratory of Environmental Change and Natural Disaster of Ministry of Education,Faculty of Geographical Science,Beijing Normal University,Beijing 100875,China;Faculty of Geographical Science,Beijing Normal University,Beijing 100875,China)

机构地区：[1]航天科工海鹰集团有限公司,北京100070 [2]北京师范大学环境演变与自然灾害教育部重点实验室,北京100875 [3]北京师范大学地理科学学部,北京100875

出　　处：《防灾减灾工程学报》2024年第1期12-27,共16页Journal of Disaster Prevention and Mitigation Engineering

基　　金：第二次青藏高原综合科学考察研究项目(2019QZKK0906);中国科学院战略性先导科技专项(XDA2003020201)资助。

摘　　要：网络灾情数据为震后损失快速评估提供了全新的视角,而网络灾情数据语义上的模糊性和内容上的冲突性严重制约了其在应急评估中的应用。为解决其模糊性与冲突性,充分挖掘其价值,构建了基于网络灾情数据融合的建筑物倒塌快速评估模型。首先,在网络灾情数据核心词汇提取的基础上,基于云模型理论,结合蒙特卡洛模拟思想,对网络灾情数据的模糊性进行量化;其次,通过网络灾情数据的可信度和模糊性度量,对冲突数据进行修正;然后,基于D-S证据理论,融合修正后的数据,获取建筑物倒塌率评估结果;最后,利用漾濞5·21 M_(s)6.4地震后15 h、玛多5·22 M_(s)7.4地震和芦山4·20 M_(s)7.0地震后12 h内的网络灾情数据,验证该模型的有效性。案例研究表明,评估结果与实际损失相吻合,该模型可为震后建筑物倒塌快速评估提供辅助支持。Web-based disaster data provides a new perspective for the rapid assessment of post-earth-quake damage.However,the semantic ambiguity and content conflict of web-based disaster data se-verely restrict its application in emergency assessment.To fully derive its value,a rapid building col-lapse assessment model based on the fusion of web-based disaster data is developed.First,on the ba-sis of the kernel vocabulary extracted from web-based disaster data,the ambiguity of web-based disas-ter data is quantified using the cloud model theory and the Monte Carlo simulation idea.Secondly,conflict data is corrected by measuring the reliability and ambiguity of the web-based disaster data.Thirdly,based on the D-S evidence theory,the modified conflict data is fused to obtain the assess-ment results of the building collapse rate.Finally,the effectiveness of the model was verified using network disaster data within 15 hours after the Yangbi M_(s) 6.4 earthquake and 12 hours after the Maduo M_(s) 7.4 earthquake and the Lushan M_(s) 7.0 earthquake.The study results show that the assessment re-sults are consistent with actual losses in the three earthquakes,which can provide auxiliary support for rapid assessment of post-earthquake building collapse.

关 键 词：网络灾情数据 建筑物倒塌 数据融合 地震灾害 

分 类 号：P954[天文地球—自然地理学] X43[环境科学与工程—灾害防治]