Settlement behavior plays an important role for the stability of underwater tunnel due to the different responses of fractured surrounding rock to external load. In contrast to the traditional analysis method based on continuum mechanics, the presented numerical model using improved hybrid finite element was performed to study the settlement behaviors of structure, and structural health monitoring system (SHMS) was introduced for field verification. The Nanjing Yangtze River tunnel, a typical underwater shield tunnel was selected as a case study for numerical simulation and real-time monitoring. First, an improved numerical model was developed on the basic of our previous research, which considers the impact of natural geological fractures on structure stability. Then, numerical investigation was applied to study the displacement of settlement under different boundary conditions. The differences between intact and fractured surrounding rock were discussed, which denote the response of fractured surrounding rock is significantly larger. To verify the numerical results, SHMS was employed in this project to monitor its mechanical behaviors. On the basis of the mass monitoring data, the analytical method was introduced to investigate the response of tunnel settlement to water pressure, which agreed well with the results obtained from the model of fractured surrounding rock.

Shield tunneling is one of the most important technologies for building of underground engineering. Many grouting holes were prefabricated for the requirement of backfill grouting, which is easy to induce local damages and potential disasters, such as leakage and cracking. Accordingly, an integrated workflow for damage detection and stability evaluation was performed based on nondestructive testing (NDT) and numerical simulation. As a case study, this method was applied to an underwater shield tunnel. Firstly, Ground Penetrating Radar (GPR) was used to detect the conditions in grouting holes. Then, the infrared camera was used to determine the damaged positions induced by grouting holes. According to NDT results, the numerical models were developed to analyze the mechanical behaviors of structure. It indicated the geophysical inversion results are consistent with field conditions. The influence area increases with a significant value of water pressure, and stress magnitude would increase to 45KPa if the increment of water pressure reaches to 10KPa. As a promising application, structure stability was evaluated in the light of analytical results.

Plasticity theory aims at describing the yield loci and work hardening of a material under general deformation states. Most of its complexity arises from the nontrivial dependence of the yield loci on the complete strain history of a material and its microstructure. This motivated 3 ingenious simplifications that underpinned a century of developments in this field: 1) yield criteria describing yield loci location; 2) associative or nonassociative flow rules defining the direction of plastic flow; and 3) effective stress-strain laws consistent with the plastic work equivalence principle. However, 2 key complications arise from these simplifications. First, finding equations that describe these 3 assumptions for materials with complex microstructures is not trivial. Second, yield surface evolution needs to be traced iteratively, i.e., through a return mapping algorithm. Here, we show that these assumptions are not needed in the context of sequence learning when using recurrent neural networks, diverting the above-mentioned complications. This work offers an alternative to currently established plasticity formulations by providing the foundations for finding history- and microstructure-dependent constitutive models through deep learning.