摘要
传统的GM(1,1)灰色理论模型,一般适用于等时距数据序列的模拟预测。本文针对数据序列本身要满足灰指数规律,当灰数据发生跨越增长的时候,采用传统的GM(1,1)模型预测精度比较差,而且从GM(1,1)模型的建模基础考虑,预测精度受初始值和背景值影响很大,由于客观条件的影响,边坡数据序列的获得有的时候不可能达到严格的等时距的数据序列,从而建立起适合边坡变形值预测的GM(1,1)模型。
Traditional GM ( 1,1 ) gray theoretical model generally applies to equal time-series data and data sequence itself should meet the gray index regulation. When index increase by leaps and bounds, the prediction accuracy of traditional GM ( 1,1 ) model is rather poor. According to the modeling base of GM ( 1,1 ) model , the forecas- ting accuracy is influenced greatly by the initial value and background value. Restricted by the objective conditions, the slope sequence data sometimes arent strict equal time-series data sequence. But the paper establishes a suitable slope deformation value forecast GM ( 1,1 ) model to solve the problem.
出处
《黄金》
CAS
北大核心
2008年第9期23-25,共3页
Gold
关键词
边坡变形
GM(1
1)
灰色
预测
slope deformation
GM ( 1,1 )
gray
forecast
low growth
high growth
作者简介
张飞(1959-),男,内蒙古包头人,教授,硕士生导师,院长,从事矿业工程与岩土工程领域的教学与科研工作;内蒙古包头市昆区阿尔丁大街7号,014010