加权非等间隔灰色Verhulst模型及其应用  

Weighted Non-equal Interval Grey Verhulst Model and Its Application

作　　者：王江荣[1] 蒲晓妮[1] 赵睿[1] WANG Jiang-rong;PU Xiao-ni;ZHAO Rui(College of Information Processing and Control Engineering,Lanzhou Petrochemical Polytechnic,Lanzhou 730060,China)

机构地区：[1]兰州石化职业技术学院信息处理与控制工程学院

出　　处：《数学的实践与认识》2019年第20期176-182,共7页Mathematics in Practice and Theory

基　　金：兰州市西固区科学技术局计划项目(西科发[2017]29号);兰州市科学技术局计划项目(兰财建发[2015]85号)

摘　　要：针对建筑物沉降监测数据量少、信息贫乏,且具非等时距、呈"S"形变化的特点,建立了能够反映饱和状态的非等间隔灰色Verhulst模型.因早期建模数据对模型预测精度贡献较小,而靠近预测时间的建模数据贡献则较大,故建模时按时间先后对建模数据赋予了不同的权重,以此体现贡献大小.实证分析表明,采用数据加权能够提高模型的预测精度,从对建模以外的监测数据预测结果来看,效果优于已有一些模型.Aiming at the characteristics of small amount of data,poor information,unequal time interval and "S" shape change of building settlement monitoring,a non-equal interval gray Verhulst model which can reflect the saturation state is established.Because the early modeling data contributes less to the prediction accuracy of the model,and the modeling data near the prediction time contributes more,therefore,the modeling data are given different weights according to the time Therefore,the modeling data are given different weights according to the time sequence,in order to reflect the contribution size.Empirical analysis shows that data weighting can improve the prediction accuracy of the model.From the prediction results of monitoring data other than modeling,the effect is better than that of some existing models.

关 键 词：沉降 非等间隔 灰色Verhulst 模型 加权 预测 

分 类 号：TP3[自动化与计算机技术—计算机科学与技术]