摘要
针对软土地基沉降周期长,承载力低,沉降变形波动大等特点,以某高铁沉降数据为依托,应用双曲线法、Asaoka法、星野法、GM(1,1)模型进行沉降预测,分析不同时间起点对预测结果的影响.结合实际沉降结果对比分析得出:①负沉降点对预测精度影响较大,沉降预测时最好跳过或删除负沉降,且不易采用负沉降点附近的数据作为参考数据;②双曲线法、Asaoka、GM(1,1)模型较适合于软土地基沉降预测,而星野法拟合效果相对较差;③双曲线法和GM(1,1)模型以停载后5个月为预测时间起点结果较准确,Asaoka法受预测时间起点影响不明显.
The soft soil foundation is featured by long deposition period, low bearing capacity, settlement of instability and large deformation. Based on a high iron settlement data, the author applied hyperbola method, Asaoka method, the hoshino method and GM （1, 1 J model for settlement prediction, and analysis the influence of different time starting point to the prediction. Conclusions can be reached by comparing four forecast methods＂ accuracy combined with real settlement results. The results show that Negative settlement site place great adverse influence on the prediction accuracy. So the author suggested when perdict settle- ment, Negative settlement sites should be deleted. The fitting effects of Hyperbolic method, Asaoka, GM （1, 1} model are good,and hoshino method is relatively poor. As for Hyperbolic method and GM （1, 1） model,prediction results are more accurate when take five month after constant-load as the starting point. The choice of different time point has little effect on the Hushino method and Asaoka method.
出处
《湘潭大学自然科学学报》
CAS
北大核心
2014年第1期38-44,共7页
Natural Science Journal of Xiangtan University
关键词
高铁路基
地基沉降
沉降预测
substructure of the high-speed rail
foundation settlement
settlement prediction
