不完备概率信息条件下浅埋软土隧道盾构施工的系统可靠性研究

Soil parameters are associated with an intrinsic uncertainty. Due to the limited experimental data, it is difficult to obtain the complete probability information on soil properties. Therefore, it is important to estimate shield tunneling reliability based on incomplete probability information. This research project attempts to construct the joint probability distribution function using the incomplete probability information on soil properties. The dependence structures of soil parameters under different construction method are investigated to identify the dispersion of the joint probability. A method for improving the accuracy of uncertainty characterization is also explored. Moreover, shield tunneling in urban area with soft soils and shallow cover depth can trigger excavation face instability and damages to the environment. It is necessary to analyze the ultimate limit state (ULS) and serviceability limit state (SLS) of shield tunneling, and describe the failure boundary under different failure mode and the system failure boundary considering coupled failure modes. The upper and lower bounds of system failure probability are then derived based on the incomplete probability information, and a method for improving the accuracy of failure probability estimation is established. By soil test, site measuring, numerical analysis, and data mining techniques, it is expected to reveal the propagation mechanism of soil uncertainties and clarify the relationships between support parameters and system reliability of shield tunneling. This research project can promote the development of risk management and disaster prevention techniques regarding shield tunneling. The research results can provide directions for shield tunneling operation.

不确定性是土体参数的固有属性，受限于试验条件，实际工程中难以获取土体参数的完备概率信息，基于不完备概率信息精确评估隧道盾构施工可靠性意义重大。本项目拟基于不完备概率信息构造土体参数的联合概率分布函数，研究不同构造方案下土体参数的相关结构，明确联合概率的波动范围，探索提高土体参数不确定性表征精度的方法。针对城市浅埋软土隧道盾构施工可能引发的开挖面失稳和周边环境破坏问题，分析盾构开挖的承载能力极限状态和正常使用极限状态，刻画不同失效模式下的失效域边界和复合失效模式下的系统失效域边界，阐明不完备概率信息条件下盾构施工的系统失效概率上下界限，建立提高失效概率估计精度的方法。通过土工试验和现场监测手段，结合数值分析与数据挖掘技术，深度揭示土体参数不确定性的传播机制，澄清支护参数与施工可靠度的影响关系。本研究可推动有关盾构施工的风险管理和防灾减灾技术的发展，相关成果可为隧道盾构施工提供科学依据。

不确定性是土体参数的固有属性，受限于试验条件，实际工程中难以获取土体参数的完备概率信息。另一方面，盾构施工存在多种失效模式，需要考虑多失效模式复合作用下的系统可靠性。本项目针对上述问题研究了不完备概率信息条件下的土体参数不确定性表征与复合失效模式下的盾构施工系统可靠性评估。界定了不完备概率信息缺失的关键要素，基于copula理论建立了随机向量联合概率分布的解析表达，提出了不完备概率信息下土体参数的藤式copula不确定性表征方法，得出了一套copula参数的序贯搜索算法。分别定义了以开挖面极限支护压力和地表沉降为指标的盾构施工承载力极限状态和适用性极限状态，基于自适应响应面和随机响应面显性表达了两种失效模式的状态方程，提出了基于重构联合概率分布的Rosenblatt变换，证明了高斯相依结构下与Nataf变换的等价性，与一/二阶可靠度法、子集模拟结合建立了高效可靠性评估方法。揭示了土体参数之间存在的下尾相关结构，阐明了适用性极限状态为盾构施工的主要失效模式，厘清了非线性相依结构对盾构施工系统可靠性的影响，与土体实际相依结构相比，常用高斯线性相依结构会高估盾构施工的系统可靠性从而导致非保守的施工方案。该研究成果提高了不完备概率信息条件下可靠性的评估精度，为地下工程施工防灾减灾提供了科学依据。