Abstract: Loess high fill subgrade is prone to uneven settlement deformation. In order to better study its settlement mechanism and establish a settlement prediction mechanism and guide specific projects to ensure project safety, this article relies on the K46+500 project of the typical loess high-filled subgrade section of the ZT5 bid section of the Yanqing to Chongli Expressway in the Winter Olympics,and use the measured data from three settlement monitoring points at the site and use the logarithmic curve fitting method and hyperbola the fitting method and the power curve fitting method used to fit and analyze the measured settlement data in the field to establish a more accurate settlement prediction model. The model- DOI:10.3969/j.issn.1000-1565.2020.05.002黄土质高填方路基沉降变形与预测杨三强¹,段士超¹,刘娜¹,吴浩楠²(1.河北大学 建筑工程学院,河北省土木工程监测与评估技术创新中心 河北 保定 071002;2.河北建设集团股份有限公司 路桥公司,河北 保定 071070)摘 要:黄土质高填方路基易发生不均匀沉降变形,为了更好地研究其沉降机理以及建立沉降预测机制,指导具体工程确保工程安全,本文依托冬奥会延崇高速公路ZT5标段典型黄土高填方路基断面K46+500工程,利用现场该断面处3个沉降监测点的实测数据,通过对数曲线拟合法、双曲线拟合法和乘幂曲线拟合法对现场实测沉降数据进行拟合对比分析,得出预测精确合理的模型.通过高填方路基数值模拟计算,分析其沉降变形特点,并对沉降预测模型进行验证.研究结果表明:1)3种曲线拟合法的拟合相关系数都在0.98以上,都能对该高填方路基短期沉降做出准确预测.2)对3种预测模型的预测结果值与实测值比较分析,得出双曲线模型的预测结果值与实测值之间的误差平方和最小为4.09 mm²,预测精度最高.3)通过建立的双曲线预测模型对高填方路基沉降进行最终沉降预测,得出路基在517 d时沉降速率为0.01 mm/d,沉降达到稳定.4)利用FLAC3D的软件建模分析对比双曲线预测模型,验证了双曲线模型的可靠性.关键词:高填方路基;黄土土质;沉降监测;预测模型中图分类号:U416.1 文献标志码:A 文章编号:1000-1565(2020)05-0454-07Settlement deformation andprediction of loess high fill subgradeYANG Sanqiang¹, DUAN Shichao¹, LIU Na¹, WU Haonan²(1.Hebei Civil Engineering Monitoring and Performance Evaluation Technology Innovation Center,College of Civil Engineering and Architecture, Hebei University, Baoding 071002, China;2.Rood and Bridge Branch,Hebei Construction Group Corporation Limited,Baoding 071070, China)Abstract: Loess high fill subgrade is prone to uneven settlement deformation. In order to better study its settlement mechanism and establish a settlement prediction mechanism and guide specific projects to ensure project safety, this article relies on the K46+500 project of the typical loess high-filled subgrade section of the ZT5 bid section of the Yanqing to Chongli Expressway in the Winter Olympics,and use the measured data from three settlement monitoring points at the site and use the logarithmic curve fitting method and hyperbola the fitting method and the power curve fitting method used to fit and analyze the measured settlement data in the field to establish a more accurate settlement prediction model. The model- 收稿日期:2019-09-28 基金项目:河北大学创新资助项目(hbu2019ss019);河北省自然科学基金资助项目(E2018201106);河北省交通运输厅科技项目(TH-201925;TH-201918);河北省高层次人才资助项目(B2017005024);交通运输部科技示范项目(2016010) 第一作者:杨三强(1980—),男,四川绵阳人,河北大学教授,博士后,主要从事公路工程路基路面工程研究.E-mail:ysq0999@163.com第5期杨三强等:黄土质高填方路基沉降变形与预测of high fill roadbed is established, the characteristics of settlement and deformation are analyzed, and the established settlement prediction model is verified. The research results show that: 1)The fitting correlation coefficients of the three curve fitting methods are all above 0.98, which can accurately predict the short-term settlement of the high-filled subgrade. 2)By comparing and analyzing the prediction results of the three prediction models and the measured values, it is found that the minimum error sum of squares between the predicted results and the measured values of the hyperbolic model is 4.09 mm², and the prediction accuracy is the highest. 3)Based on the established hyperbolic prediction model, the final settlement prediction of the settlement of high-filled subgrade is carried out, and it is concluded that the settlement rate of the subgrade is 0.01mm/d at 517 days, and the settlement is stable. 4)By using FLAC3D software to analyze and compare the hyperbolic prediction model to verify the reliability of the hyperbolic model.

Key words: high fill subgrade, the loess, settlement observation, predictive model