摘要
针对传统GM(1,1)模型存在数据序列的初始值过旧使预测意义减弱的问题,文中采用更新数据序列初始值的新陈代谢GM(1,1)模型对传统模型进行改进;以南方某大坝边坡监测点的沉降位移为例,分别使用两种模型对该监测点沉降位移进行拟合预测并与实际值进行比较。实验证明,新陈代谢GM(1,1)模型精度明显高于传统模型,更接近于真实值。
It introduces the grey prediction GM(1,1)model,because the traditional GM(1,1)model's initial value of the data sequence is too old to diminish defects such as the meaning of forecast.The metabolism GM(1,1)model is used,to update data sequence of the initial value to improve the traditional model.Through a dam slope monitoring point's settlement displacement data in the south of China,separately two models are proposed to forecast the settlement displacement of this monitoring point,fited and compared with the actual value.The experiment proves that,the metabolic GM(1,1)model precision is significantly higher than the traditional model,more closely to the real value.
出处
《测绘工程》
CSCD
2015年第8期53-56,共4页
Engineering of Surveying and Mapping
基金
中央高校基本科研业务经费资助项目(2012B01714)
