摘要
建筑物沉降监测数据往往满足时间序列等时性的要求,对观测数据进行平稳化处理后,可以建立时序模型。该文利用已有的沉降观测数据,建立ARMA模型,并对建筑物沉降量进行预测,取得了较好的拟合效果和预测精度。
When settlement observation data for buildings has characteristics of equal lag, it can be transformed into stationary time series to set up time series model. This paper proposes a trend prediction method for settlement by setting up ARMA model, The model can fit settlement time series quite well and achieve better prediction accuracy.
出处
《勘察科学技术》
2006年第6期52-54,共3页
Site Investigation Science and Technology
