摘要
本文以高层建筑物沉降变形预测为主要研究目的,讨论了GM(1,1)方法适用于单一指数增长模型、对预测序列数据异常情况难以准确预测的局限性,利用线性回归适用短期预测的特点,提出了基于GM(1,1)与线性回归组合预测高层建筑物沉降变形的方法;对组合模型预测精度起决定性作用的灰指数v和参数m进行了分析,给出了求解灰指数v和参数m的最优值算法,最后利用组合模型对某高层建筑物沉降变形数据进行了解算,应用结果表明,该方法使预测结果更为可靠、准确。
Regarding forecasting the subsidence and deformation of high building as the main research purpose, the paper discussed the limitation of GM ( 1, 1 ) model in forecasting accidents with simple index increase model which had some difficult in dealing with the abnormity circs of list data. Using the character of linear regression adapting to the short term forecast, a combined method based on GM ( 1, 1 ) and line regression was presented, and grey parameter v and parameter m were analyzed on how they affect the results of fitting. According to the experimental results, the method could improve the prediction accuracy.
出处
《测绘科学》
CSCD
北大核心
2012年第3期96-98,共3页
Science of Surveying and Mapping
基金
国家自然科学基金项目(41071328)
国家重点基础研究发展规划(973)项目(2007CB209400)
教育部新世纪优秀人才支持计划资助项目(NECT-07-07098)
矿山空间信息技术国家测绘局重点实验室开放基金资助项目(KLM200816)
关键词
高层建筑物
沉降变形
GM(1
1)
线性回归
灰指数v
参数m
预测
high building
subsidence and deformation
GM (1, 1)
linear regression
grey parameter v
parameter m
prediction