期刊文献+

基于大数据分析的火灾预警方法研究 被引量:4

Research on Fire Early Warning Method Based on Big Data Analysis
原文传递
导出
摘要 城市中一旦发生火灾将伴随人员伤亡及财产损失,对于城市火灾的发生进行准确预测并采取有效控制措施具有重要重要意义.传统方法存在精度低、计算耗时长等缺点.采用Bootstrap方法、二项分布的方法结合大数据分析并采用超级计算机运行速度较快的优势,可发现数据隐含关系,识别特征数据,提高火灾预测精度.利用天津市2015-2017年火灾数据,采用Bootstrap方法、二项分布的方法构造相应的置信水平为95%的预测区间.根据2015-2016年火灾数据,用2017年的火警数量进行验证,并选取精度较高27%以上的权重作为预测部分中的权重.利用2015-2017年的数据,通过Bootstrap方法、二项分布的方法预测2018年同期情况,得出预测结果吻合度大于63.33%.该模型可应用于火灾预测,其结果可为消防部门制定消防措施及理论依据. If a fire breaks out in a city,it will cause casualties and property losses,It is of great significance to accurately predict the occurrence of urban fires and take effective control measures.The traditional method has some shortcomings,such as low accuracy and long calculation time.Bootstrap method and binomial distribution method combined with large data analysis and the advantage of supercomputer running faster,The implicit relationship of data can be found,the characteristic data can be identified,and the accuracy of fire prediction can be improved.Based on the fire data of Tianjin from 2015 to 2017,the prediction interval with 95%confidence level is constructed by Bootstrap method and binomial distribution method.Based on the fire data from 2015 to 2016,the number of fire alarms in 2017 is used to verify,and the weight with a higher accuracy of 27%is selected as the weight of the prediction part.Based on the data of 2015,2016 and 2017,the same period in 2018 is predicted by Bootstrap method and binomial distribution method,The consistency of the predicted results is more than 63.33%.The model can be applied to fire prediction,and the results can be used to formulate fire control measures and theoretical basis for fire departments.
作者 李庆功 王悦 李继繁 高佑强 宋文华 Li Qinggong;Wang Yue;Li Jifan;Gao Youqiang;Song Wenhua(Tianjin Fire Rescue Bureau,Tianjin 300090,China;School of Control and Mechanical Engineering,Tianjin Chengjian University,Tianjin 300384,China;School of Environmental Science and Engineering,Tianjin Polytechnic University Tianjin 300387,China;National Supercomputer Center in Tianjin,Tianjin 300000,China)
出处 《南开大学学报(自然科学版)》 CAS CSCD 北大核心 2020年第4期108-112,共5页 Acta Scientiarum Naturalium Universitatis Nankaiensis
基金 天津市科技重大专项与工程(16ZXHLSF00290)。
关键词 火灾预测 BOOTSTRAP 二项分布 fire prediction Bootstrap binomial distribution
  • 相关文献

参考文献8

二级参考文献63

共引文献50

同被引文献36

引证文献4

二级引证文献13

相关作者

内容加载中请稍等...

相关机构

内容加载中请稍等...

相关主题

内容加载中请稍等...

浏览历史

内容加载中请稍等...
;
使用帮助 返回顶部