多元回归分析方法在高层建筑沉降监测数据处理中的应用

Application of Multiple Regression Analysis in Data Processing of High-rise Buildings Settlement Monitoring

高层建筑物的监测周期一般较长,数据量大,数据处理和分析较为复杂,因此,需要选择一种较为合适的、贴近工程实际的分析方法。本文以某高层建筑物的沉降监测数据为例,阐述了多元回归分析模型中影响因子的确定、回归模型的最小二乘参数估计以及回归模型的显著性检验,并比较了沉降累计值的预测值和真实值,结果显示:二者的残差较小,可见预测精度较高,说明多元回归分析在高层建筑沉降监测数据处理中是行之有效的。

The monitoring period of high - rise buildings are usually quite long and there are a large number of data to process, whose processing and analysis are relatively complicated. Hence a proper analytical method that is close to the engineering reality must be chosen. Taking the data processing of one high - rise building as an example, this paper explains the determination of impact factors in multiple regression analysis models, the least square parameters estimation and significance test of regression model, and a comparison between the predicted value and real value of the lost settlement is also made. The result shows that the residual between them are quite small, so it can be seen that the prediction accuracy of multiple regression analysis is very high and its use in data processing of high -rise buildings settlement monitoring is effective.