灰色预测与Kalman滤波在建筑物沉降变形分析中的应用  

Application of grey prediction and Kalman filter in building settlement deformation analysis

作　　者：周吕[1] 文鸿雁[1] 韩亚坤[1] 

机构地区：[1]桂林理工大学测绘地理信息学院/广西空间信息与测绘重点实验室,广西桂林541004

出　　处：《测绘科学》2014年第4期149-151,82,共4页Science of Surveying and Mapping

基　　金：国家自然科学基金项目(41071294);广西空间信息与测绘重点实验室资助课题(桂科能1103108-02);广西科学基金资助项目(桂科基0991023);广西研究生教育创新计划资助项目(YCSZ2012083)

摘　　要：沉降变形分析与预测对于建筑物的安全运营具有重要作用,建立科学、合理的预测模型对于变形分析极其重要。本文运用GM(1,1)模型与Kalman滤波模型对建筑物沉降变形进行预测。通过实例对比分析表明,GM(1,1)模型较适合短期且变形趋势呈线性或指数分布趋势的变形分析与预测;Kalman滤波模型对短期和较长周期呈波形或线性变形均具有较高预测精度。Settlement deformation analysis and prediction play an important role in the safe operation of building, establishing scientific and reasonable prediction model is extremely important in deformation analysis. This paper used GM （1, 1） model and Kalman filter model to predict the settlement deformation of building. Through comparing and analyzing the projects it was shown that GM （1, 1） model could be more suitable for short-term deformation analysis and prediction when the deformation trend is linear or exponential distribution, while Kalman filter model would show a higher prediction precision in short-term and longer period waveform or linear deformation.

关 键 词：沉降观测 KALMAN滤波 GM(1 1) 变形预测 

分 类 号：TU196[建筑科学—建筑理论]