首页 | 本学科首页   官方微博 | 高级检索  
     检索      

基于DBSCAN聚类的城市区域火灾风险计算方法——以深圳市盐田区为例*
引用本文:王倩倩,孟繁宇,曾益萍,张少标,吴国华,杨丽丽.基于DBSCAN聚类的城市区域火灾风险计算方法——以深圳市盐田区为例*[J].中国安全生产科学技术,2021,17(2):177-182.
作者姓名:王倩倩  孟繁宇  曾益萍  张少标  吴国华  杨丽丽
作者单位:(1.南方科技大学 统计与数据科学系,广东 深圳 518055;2.南方科技大学 前沿与交叉科学研究院,广东 深圳 518055;3.深圳市城市公共安全技术研究院,广东 深圳 518055)
基金项目:* 基金项目: 国家重点研发计划项目(2018YFC0807000,2019YFC0810700);国家自然科学基金项目(71771113)
摘    要:为探究评估城市区域的火灾综合风险,提出空间聚类和层次分析法对深圳市盐田区2011年至2019年火灾数据进行建模分析。运用DBSCAN聚类模型对火灾地点进行分类,结合层次分析法构建火灾风险评估验证模型,探讨聚类结果、火灾场景、火灾原因构建系统层次结构模型,综合得出风险评估结果。通过将聚类区域还原至百度地图,结合场景信息及层次分析法指标值对模型的有效性进行验证。结果表明:由于各区区域功能及地理位置不同,区域火灾风险值有明显差异,模型计算出的4个区域结果与实际火灾情况具有一致性。

关 键 词:区域火灾风险  DBSCAN聚类算法  层次分析法

Calculation method of fire risk for urban areas based on DBSCAN clustering:A case study of Yantian District,Shenzhen
WANG Qianqian,MENG Fanyu,ZENG Yiping,ZHANG Shaobiao,WU Guohua,YANG Lili.Calculation method of fire risk for urban areas based on DBSCAN clustering:A case study of Yantian District,Shenzhen[J].Journal of Safety Science and Technology,2021,17(2):177-182.
Authors:WANG Qianqian  MENG Fanyu  ZENG Yiping  ZHANG Shaobiao  WU Guohua  YANG Lili
Institution:(1.Department of Statistics and Data Science,Southern University of Science and Technology,Shenzhen Guangdong 518055,China;2.Academy for Advanced Interdisciplinary Studies,Southern University of Science and Technology,Shenzhen Guangdong 518055,China;3.Shenzhen Urban Public Safety and Technology Institute,Shenzhen Guangdong 518055,China)
Abstract:In order to explore and assess the comprehensive fire risk of urban areas,the modeling and analysis on the fire data of Yantian district,Shenzhen from 2011 to 2019 were conducted by using the spatial clustering and analytic hierarchy process (AHP) methods.The DBSCAN clustering model was used to classify the fire location,and combined with the analytic hierarchy process method,a fire risk assessment and verification model was constructed,and the clustering results,fire scenarios and fire causes were discussed to establish a hierarchical structure model of the system,then the risk assessment results were comprehensively obtained.The effectiveness of the model was verified by restoring the clustering area to Baidu map combined with the scene information and index values of analytic hierarchy process.The results showed that due to different regional functions and geographic locations of each area,the regional fire risk values were significantly different,and the four regional results calculated by the model were consistent with the actual fire situation.
Keywords:regional fire risk  DBSCAN clustering algorithm  analytic hierarchy process
本文献已被 CNKI 等数据库收录!
点击此处可从《中国安全生产科学技术》浏览原始摘要信息
点击此处可从《中国安全生产科学技术》下载免费的PDF全文
设为首页 | 免责声明 | 关于勤云 | 加入收藏

Copyright©北京勤云科技发展有限公司  京ICP备09084417号