摘要
首先对建筑物沉降数据序列进行了平稳化处理,然后研究了平稳化序列的建模和预报方法,最后结合建筑物沉降监测的具体实例进行了时间序列的分析与预报.结果表明:将时间序列分析方法应用于建筑物沉降监测,具有建模容易、计算简单、预报快速的特点;时间序列分析方法对建筑物沉降具有较高的模型拟合及预报精度,尤其是短期预报,效果更佳;应尽量避免使用时间序列进行中长期预报,要根据实测数据对所建模型进行实时更新.
With the original data of building settlement being processed smoothly, a study was performed of the process of modeling for prediction of stationary series. Time series analysis and prediction in a case study of settlement monitoring of buildings show that the introduction of time series analysis method into settlement monitoring of buildings makes it easy for modeling, computation and realization of fast prediction, and that it has high precision for model fitting and predication of building settlement, especially for short-term prediction. However, the time series is ineffective for middle and long-term prediction, and the developed model should be updated according to the measured data.
出处
《河海大学学报（自然科学版）》
CAS
CSCD
北大核心
2006年第4期426-429,共4页
Journal of Hohai University(Natural Sciences)
关键词
建筑物
沉降
时间序列分析
趋势项
平稳序列
预报
building
settlement
time series analysis
tendency item
stationary series
prediction
