非负变权组合预测模型在建筑物沉降预测中的应用  

Application of Variable Weighting Combination Forecasting Model to Building Subsidence

作　　者：谌芳 CHEN Fang(School of Geomatics,Anhui University of Science and Technology,Huainan Anhui 232001,China)

机构地区：[1]安徽理工大学测绘学院,安徽淮南232001

出　　处：《北京测绘》2018年第4期447-451,共5页Beijing Surveying and Mapping

摘　　要：为了克服建筑物变形预测中单一模型预测精度差、不稳定的问题,本文融合了适应性强、互补性好的自回归滑动模型、灰色模型和三次指数平滑法三种模型,基于误差平方和最小准则,采用滚动时间域的方式构建了非负变权组合预测模型。工程应用表明:非负变权组合预测模型的预测精度、可靠性优于三种单一预测模型、方差倒数组合预测模型以及等权组合预测模型。研究成果对建筑物沉降预测具有较好的参考价值。In order to overcome the poor prediction accuracy and instability of single model in building deformation prediction,This paper combines three models of AR model,GM(1,1)model and cubic exponential smoothing method which with strong adaptability and good complementarity,we propose an optimal non-negative variable weight combination forecasting model based on the error square of minimum.Engineering application shows:The prediction accuracy of non-negative variable weight combination forecasting model is obviously better than that of AR model,GM(1,1)model,cubic exponential smoothing model and combination forecasting model based on reciprocal of variance and equal weight.The study results may be available for reference to settlement prediction.

关 键 词：自回归滑动模型(AR) 灰色模型(GM) 三次指数平滑法 变权 建筑物 

分 类 号：P258[天文地球—测绘科学与技术]