论文部分内容阅读
在深基坑工程数值分析中,土体本构关系的选取及其参数的确定对分析的结果会产生重要的影响。结合一个海中深厚软土基坑实例,根据现场实测数据,利用参数敏感性分析确定地层的关键参数,提出一个基于参数敏感性分析和BP神经网络的动态施工反演分析方法。分析结果显示,围护桩水平位移监测值与反演值吻合较好,说明基于参数敏感性分析和BP神经网络的动态施工反演分析方法能够很好地反映土体的非线性、不确定特征,依据该反分析方法获得的土体参数可用于估算下一工况的基坑位移。本次反演得到的土体参数可作为当地深厚软土地层类似工程土体参数的参考取值。
In the numerical analysis of deep foundation pit engineering, the selection of soil constitutive relationship and the determination of its parameters will have a significant impact on the analysis results. Combined with a case study of a deep deep foundation pit in the sea, the key parameters of the formation are determined by parameter sensitivity analysis based on field measurements. A dynamic construction inversion method based on parameter sensitivity analysis and BP neural network is proposed. The analysis results show that the monitoring value of the horizontal displacement of the retaining pile is in good agreement with the inversion value, which shows that the dynamic construction inverse analysis method based on parameter sensitivity analysis and BP neural network can well reflect the nonlinear and uncertain characteristics of soil , The soil parameters obtained from this back analysis method can be used to estimate the displacement of foundation pit in the next working condition. The soil parameters obtained by this inversion can be used as the reference values of soil parameters of similar projects in deep deep soft soil formations.