基于椭圆拟合的隧道断面监测及其应用

摘要

On the basis of least square method, an improved method is disclosed in this paper. Firstly, conducting least-square fitting, then using the sample median to eliminate the gross errors in data, finally, the least-square algorithm is used to fit the elliptic curve. This method could enhance the robustness of the fitting results. Here, each of the deformation is calculated by high-precision ellipse fitting, and the convergence ratio is of great response to the trend of the micro-deformation of tunnel cross-section. Moreover, the data of Xinmofan Road tunnel is used to validate this method, and the result is good.%基于最小平方中值法提出了一种改进的方法,即先进行最小平方中值法拟合,再用样本中位数剔除数据中的粗差,最后用最小平方中值算法拟合椭圆曲线,增强了拟合结果的稳健性.通过高精度的椭圆拟合,可以计算出每次的变形量,且提出的收敛比例很好地反映了隧道断面某一部分的微变形趋势.利用新模范马路隧道的监测数据进行了验证.