基于灰色分析的土石堤坝地基沉降预测方法

Prediction method of soil embankment settlement based on grey analysis

目前提出的土石堤坝地基沉降预测方法难以对离散数据进行拟合分析,导致地基沉降预测结果 误差较大,残差校正效果较差。为了解决上述问题,基于灰色分析研究了一种新的土石堤坝地基沉降预测方法。分析曲线变化规律,通过沉降拟合模型分析沉降量系数,实现沉降离散数据拟合预测。建立沉降地基有限元模型,计算有限元分析参数,根据地基结构数据与沉降物理参数得到数据显性特征,通过原始沉降数据序列计算沉降灰色分析方程式,建立灰色分析模型,检测数据序列特征,计算加权系数的平方和,去除异常数据,实现沉降预测。实验结果表明,基于灰色分析的土石堤坝地基沉降预测方法误差值低于1.25%,与实际值吻合度较高,经过残差修正后拟合的平均相对误差低于0.08%,具有极高的预测能力。

It is difficult to fit the discrete data in the proposed prediction methods of soil embankment settlement,which leads to large errors in the prediction results of foundation settlement and poor residual correction effect.In order to solve the above problems,a new method of soil embankment settlement prediction based on grey analysis is studied.The variation rule of the curve was analyzed,and the settlement coefficient was analyzed by the settlement fitting model to realize the fitting prediction of discrete settlement data.The finite element model of settlement foundation is established,the finite element analysis parameters are calculated,and the dominant characteristics of the data are obtained according to the foundation structure data and settlement physical parameters.The grey analysis equation of settlement is calculated through the original settlement data series,and the grey analysis model is established to detect the characteristics of data series,calculate the sum of squares of weighted coefficients,remove abnormal data,and realize settlement prediction.The experimental results show that the error value of the prediction method based on grey analysis is less than1.25%,which is highly consistent with the actual value.The average relative error of the fitting method after residual correction is less than 0.08%,which has high prediction ability.