粗泛化箱型图法在大中跨径桥梁监测异常值剔除的应用

Application of Rough Generalization Box Plot in Eliminating Outlier of Bridge Monitoring

以东莞某营运旧桥监测项目为背景,通过增大箱型图法的系数k,使箱型图法粗泛化,并结合拉依达准则对桥梁监测数据的异常值进行识别和剔除。该方法不仅较好地解决了箱型图对异常值过度剔除的情况,也一定程度地减少了运算次数,提高了计算效率。对于未安装桥梁健康监测系统的大、中跨径旧桥监测数据异常值的剔除具有一定的工程实用价值,同时可作为桥梁健康监测系统中海量监测数据异常值剔除的一种补充方法。

Based on the monitoring project of an old operating bridge in Dongguan,in this paper,the box plot method has been generalized by increasing the coefficient k of the box plot,and the outliers of the bridge monitoring data have been identified and eliminated in combination with the Pauta criterion.This method can not only solve the problem of excessive removal of outlier from the box plot,but also reduce the number of operations to a certain extent,which can improve the calculation efficiency.It has certain engineering practical value for eliminating the outlier of the monitoring data of the old bridge with large and medium spans without installing the bridge health monitoring system,and can also be used as a supplementary method for eliminating the outlier of the massive monitoring data in the bridge health monitoring system.