基于小波去噪的地铁沉降监测分析

The settlement analysis of MTR monitoring based on wavelet

地铁等大型工程施工过程引起的地表及其周边建筑物的变形及变形趋势分析,对于确保城市及工程安全极为重要。研究有关沉降监测方法及可靠性分析,估计地表及其周边建筑物可能产生的变形,对周边建筑物的安全,以及工程本身顺利实施有着重要的现实意义。本文将沉降观测数据序列作为带有白噪声数字信号,采用小波多孔算法进行了监测数据的去噪处理,取得了较好的效果。同时论文提出采用分段拟合的方法进行数据拟合,用对数函数曲线来描述土体沉降较快的第二阶段变形趋势,变形预报较为准确。

More attentions are paid for the settlement in ground and of buildings caused by subway construction. There are important significant that the extent of the settlement in ground and of buildings around the subway sites monitored and analysed . In order to denoise data based on wavelet, the settlement data are taken as an input signal sequence in which contains Gaussian noises. The A'trous algorithm about wavelet are used in denoising the monitoring data. There are three phases used in the building settlement analysis. The logarithmic function has been used for approximating and predicting the current of building settlement,in phase of rapid subsidence of soil. The results show that the approaches given in our research projects is very perfect.