模糊聚类方法应用于大坝变形监测资料分析

Fuzzy Clustering Analysis Applied to Dam Deformation Monitoring Data

采用某混凝土重力坝2011年和2012的变形监测数据,应用模糊聚类分析对大坝变形监测点进行分类,以变形允许差的聚类效果评价方法确定最佳分类;比较分类结果,判定关键的变形监测点;考虑上、下游水位,温度,时效等因素,对关键点建立多元逐步回归统计模型,进行大坝变形预报拟合。分析结果表明,通过该方法可以高效地对大坝变形监测点的位移进行分析预报,预报结果较为精确。

Based on the deformation monitoring data of a concrete gravity dam in 2011 and 2012, the fuzzy cluster analysis is used to classify dam deformation monitoring point, and the best classification is determined by clustering effect evaluation method which taking allowed permissible error as evaluation index. The classification is compared and then the critical deformation monitoring points are finally determined. Through considering upstream and downstream water levels, temperature and aging, the multiple stepwise regression model is set up to fit and forecast the deformation of critical points. The results show that this method can efficiently analyze and forecast the displacement of dam deformation monitoring point, and the forecasting results are more accurate.