双因子模型在黄土高填方路基沉降预测中的研究分析

Research and Analysis of Settlement Prediction of High Fill Subgrade in Loess Based on Two-factor Model

选取黄土土质的高填方路基工程为研究对象,在建立沉降预测模型过程中同时考虑填土时间、填土高度双因子对沉降的影响。将常用的时间函数模型和高度函数模型进行组合对比发现,对于压实度为0.93~0.95的黄土高填方工程,时间函数采用双曲线函数、高度函数采用线性函数时双因子拟合函数的最大复杂相关系数R^2为0.931,但是由于精度不能满足工程实际需求。经过对时间函数的a_H进行调整得到了适合黄土高填方路基预测的模型,R^2为0.938,拟合度较高,且精度满足要求。该新型模型既可用于对单一点在填方高度确定情况下的预测,也可用于多点在时间跨度上的预测。

The loess soil-based high-fill embankment project is selected as the research object.In the process of establishing the settlement prediction model,the effects of the time and the height of the filling factor on the settlement are taken into account.Comparing the commonly used time function model with the height function model,it was found that for the loess high fill project with a compaction degree of 0.93 to 0.95,the hyperbolic function is used as the time function,and the maximum of the two-factor fit function is used when the height function is a linear function.The complex correlation coefficient is 0.931,but because the accuracy can not meet the actual needs of the project.After adjustment of the time function,a model suitable for loess high-fill embankment prediction was obtained with a probability of 0.938,a high degree of fit,and an accuracy that meets the requirements.This new model can be used both for the prediction of a single point in the case of fill height determination and for the prediction of multiple points over time.