1-AGO-GM(1,1)模型的构建及其在火灾死亡人数预测中应用  

Construction of 1-AGO-GM (1,1) Model and Its Application in the Prediction of Fire Death

作　　者：杨坦[1] 蒋亚龙[1] 孙志豪 YANG Tan;JIANG Yalong;SUN Zhihao(College of Civil and Environmental Engineering,Anhui Xinhua University,Hefei 230088,China;Hefei Frist Construction Engineering Co.,Ltd., Hefei 230011,China)

机构地区：[1]安徽新华学院土木与环境工程学院,安徽合肥230088 [2]合肥建工第一建筑工程有限责任公司,安徽合肥230011

出　　处：《常州大学学报（自然科学版）》2019年第4期26-30,共5页Journal of Changzhou University：Natural Science Edition

基　　金：安徽新华学院科研团队研究项目(2016td012);安徽省大学生创新创业训练计划项目(AH201512216014)

摘　　要：针对传统GM(1,1)预测模型对初始值依赖程度较高,预测结果波动性大的问题,采用一次累加方法对初始值进行修正,构建了1-AGO-GM(1,1)模型.并以北京市2008—2017年火灾死亡人数预测分析为实例,进行相应的拟合运算,对比两种模型的运算误差,结果表明改进型GM(1,1)预测精度更高,能够取得良好的预测效果.Aiming at the problem that the traditional GM(1,1) prediction model is highly dependent on the initial value and the volatility of the prediction result is large, the initial value is corrected by an accumulative method, and the 1-AGO-GM(1,1) model is constructed. Taking the prediction of fire deaths in Beijing from 2008 to 2017 as an example, the corresponding fitting calculations are carried out to compare the operational errors of the two models. The results show that the improved GM(1,1) has higher prediction accuracy,which can achieve good predictive results.

关 键 词：一次累加法 GM(1 1)模型 预测误差 

分 类 号：X913.4[环境科学与工程—安全科学]