摘要
针对天津地区地质条件复杂、软土分布广泛等问题,该文结合地铁5、6号线基坑工程实例,在现场监测成果统计的基础上,先利用皮尔逊相关系数法、极差分析法分析各因素与基坑侧向位移的相关性及敏感性;然后利用多种算法优化支持向量机的模型参数,进而构建出基坑侧向位移的最优预测模型,以期实现基坑变形的高精度预测。实例研究表明:影响天津地铁基坑变形的因素主要包括开挖深度、软土层厚度等七个因素,各因素与基坑侧向位移均处于高度相关或显著相关,且各影响因素对基坑侧向位移的敏感性存在一定差异;通过多步递进优化,提高了预测精度,验证了各类优化模型的有效性。
Aiming at the problems of complex geological conditions and wide distribution of soft soil in Tianjin,combined with the example of foundation pit enigneering of metro lines 5 and 6,based on the statistics of field monitoring results,the correlation and sensitivity between various factors and lateral displacement of foundation pit are analyzed by Pearson correlation coefficient method and range analysis method;Then,a variety of algorithms are used to optimize the model parameters of support vector machine,and then an optimal prediction model of foundation pit lateral displacement is constructed in order to realize the high-precision prediction of foundation pit deformation.The case study shows that the factors affecting the deformation of subway foundation pit in Tianjin mainly include seven factors,such as excavation depth,soft soil layer thickness and so on.Each factor is highly or significantly correlated with the lateral displacement of foundation pit,and the sensitivity of each influencing factor to the lateral displacement of foundation pit is different;Through the multi-step progressive optimization,the prediction accuracy is improved,and the effectiveness of various optimization models is verified.
作者
王飞
Wang Fei(Shaanxi Railway Institute)
出处
《勘察科学技术》
2021年第6期29-33,64,共6页
Site Investigation Science and Technology
关键词
地铁基坑
软土
侧向位移
影响因素
变形预测
subway foundation pit
soft soil
side displacement
influencing factors
deformation prediction
