摘要
为对软岩高路堤沉降变形进行合理预测,以云南某高速公路炭质页岩填料高填路堤段为依托,引入三点法取点原则对传统沉降预测方法进行改进,分析对比了改进模型与传统模型对炭质页岩高填路堤沉降变形的预测效果。选取改进模型中时间起点与时间间隔进行动态沉降预测分析,讨论了不同选点范围对改进模型预测精度的影响。结果表明:炭质页岩类软岩高填路堤沉降具有变形量大、数据波动性大的特点,使用传统模型对其进行沉降预测误差较大;基于三点修正的双曲线法与指数曲线法能有效避免数据波动带来的影响,具有预测精度高、拟合性好、计算简便的特点,能够达到较好的预测效果,而改进后的三点-星野法对沉降预测精度并无明显改进;随着模型选取时间起点与时间间隔的不同,沉降预测精度也发生较大变化,随时间起点增大,沉降量预测精度提高,相关系数可达0.99,随时间间隔增大,沉降量预测精度先增大后小幅减小,最佳选取时间间隔为56~62 d;选取的三点应能反映实测沉降数据曲线的发展趋势,当时间起点或时间间隔选取较小时,最终沉降量出现负值,误差平方和较大,拟合精度差,在对类似工程进行沉降预测时,可结合三点-双曲线法与三点-指数法进行参数试探以得到更好的预测效果。
To reasonably predict the settlement deformation of high embankment in soft rock, relying on the high-filled embankment section of an expressway with carbonaceous shale in Yunnan Province, the traditional settlement prediction method is improved by introducing three-point method, the prediction effects of the improved model and the traditional model on the settlement deformation of the high-filled embankment with carbonaceous shale are analyzed and compared. Selecting the time starting point and time interval in the improved model for dynamic settlement prediction analysis, the influence of different ranges of selected points on the prediction accuracy of the improved model is discussed. The result shows that(1) The high-filled embankment settlement of carbonaceous shale has the characteristics of large deformation and large data fluctuation, and the traditional model has a large error in settlement prediction.(2) The hyperbolic method and exponential curve method based on three-point modification can effectively avoid the influence of data fluctuation, they have the characteristics of high prediction accuracy, good fitting and simple calculation, and can achieve better prediction result, while the improved three-point Hoshino algorithm does not significantly improve the accuracy of settlement prediction.(3) With the change of time starting point and time interval selected by the model, the settlement prediction accuracy also changes greatly. As the time starting point increases, the settlement prediction accuracy improves, and the correlation coefficient can reach 0.99. As the time interval increases, the settlement prediction accuracy increase first and then decrease slightly, the best time interval is 56-62 d.(4) The selected three points should reflect the development trend of the measured settlement data curve. When the starting point or time interval is small, the final settlement is negative, the sum of squares of error is large, and the fitting effect is poor. In the settlement prediction of similar projects, the three-point hyperbolic method and three-point exponential method can be combined to test the parameters to get a better prediction effect.
作者
常洲
程小强
张留俊
徐合清
黄亚飞
CHANG Zhou;CHENG Xiao-qiang;ZHANG Liu-jun;XU He-qing;HUANG Ya-fei(CCCC First Highway Consultants Co.,Ltd.,Xi’an Shaanxi 710075,China;School of Highway,Chang'an University,Xi'an Shaanxi 710064,China;Shaanxi Key Laboratory of Highway Traffic Disaster Prevention,Xi'an Shaanxi 710075,China)
出处
《公路交通科技》
CAS
CSCD
北大核心
2022年第5期33-40,共8页
Journal of Highway and Transportation Research and Development
基金
国家重点研发计划项目(2016YFC0802203)。
关键词
道路工程
三点修正曲线模型
动态分析
软岩路堤
沉降预测
选点原则
road engineering
three-point modified curve model
dynamic analysis
soft rock embankment
settlement prediction
point selection principle
