摘要
城市空间是火灾灾害系统的载体,街区是城市管理的基本单元,火灾风险评估则是火灾风险管控的关键。本文先利用遥感图像识别结合支持向量机算法及K-means聚类获取街区尺度的空间数据,进行“单位-街区-区域”层次的高密度建成区火灾危险性评估;然后通过POI(Point of Interesting)数据挖掘,识别城市空间用地类型衡量火灾损失,对高密度建成区进行火灾风险综合评估;最后利用聚类和异常值分析探究火灾风险空间分布,提出高密度建成区火灾风险防范措施。结果表明:研究区域的539个街区单元中,有37个为五级火灾危险性街区单元,45个为五级火灾风险街区单元;街区特征暴露性是影响火灾危险性的主要因素,工业和商业活动范围是城市火灾的高损失区域;提出了集“重点监测点-重点监测区-重点改造区-消防优化区-改造优化区”于一体的石家庄市中心城区火灾风险防范建议;POI数据及遥感图像识别技术的引入,不仅提升了风险溯源能力,还弥补了火灾风险评估模型没有充分考虑城市动态变化的局限。街区尺度的火灾风险研究为高密度建成区消防安全保障提供了新思路。
Urban space is the carrier of the fire disaster system,blocks are the basic unit of urban management,and fire risk assessment is the key to fire risk management and control. First,this paper uses remote sensing image recognition technology,combined with support vector machine algorithm and K-means clustering method to obtain block-scale spatial data. Conduct fire hazard assessment in high-density built-up areas at the "unit-block-region" level. Then through POI data mining,the type of urban space land is identified to determine the fire loss coefficient,and the fire risk of high-density built-up areas is comprehensively assessed. Finally,clustering and outlier analysis are used to explore the spatial distribution of fire risks,and fire risk prevention measures for high-density built-up areas are proposed. The results show that among the 539 block units in the study area,37 are block units with level 5 fire hazard,and 45 are level 5 fire risk block units. The exposure of block characteristics is the main factor affecting fire risk,and the exposure of block characteristics is the main factor that affects the fire hazard of block. The moving area of industrial and commercial activities is the high loss area of the city city fire. Proposed fire risk prevention measures in the downtown area of Shijiazhuang,named as "Key Monitoring Points-Key Monitoring Areas-Key Reconstruction Areas-Fire Optimization Areas-Reconstruction and Optimization Areas". The introduction of POI data and remote sensing image recognition technology not only improves the traceability of fire risks,but also makes up for the limitation of the fire risk assessment model that does not fully consider the dynamic changes of the city. Conducting block-scale fire risk research can provide new ideas for improving the fire safety protection capabilities of high-density built-up areas.
作者
李凡
薛晔
李丽石
陈志芬
LI Fan;XUE Ye;LI Lishi;CHEN Zhifen(School of Economics and Management,Taiyuan University of Technology,Taiyuan 030024,China;Shijiazhuang Fire Rescue Detachment,Shijiazhuang 050000,China;China Academy of Urban Planning and Design,Beijing 100037,China)
出处
《自然灾害学报》
CSCD
北大核心
2022年第1期92-107,共16页
Journal of Natural Disasters
基金
教育部人文社会科学研究规划基金项目(20YJAZH116)
石家庄市公安消防支队委托项目(183080109-J)
国家自然科学基金项目(71901159)。
