Under the working condition of mud cake, the continuous action between the shield cutter head and the soil on the excavation surface can generate high temperature in the process of shield tunneling and excavation, which changes the characteristic of the adhesion between the soil and the cutter head to intensify the phenomenon of making a mud cake on cutter, finally leading to a vicious circle. To study the effect of temperature on the characteristics of the adhesion of the soil on the surface of the structure, the soil adhesive situation and adhesion force at different interface temperatures were tested through a self-made experiment device. According to the result, it was indicated that the moisture content has a significant effect on the adhesion force of the soil, and the adhesion force firstly increased and then decreased with the increased of the moisture content and reached the maximum value near the plastic limit moisture content. The adhesion force changes very gently when the interface temperature is low. When the temperature reached 50°C, the adhesion force continues to increase as the interface temperature continuously increases except for the soils with high moisture content; moreover, the interface temperature has a great influence on the content of soil adhered on the structure surface. As for the soil with moderate moisture content (%), this content of the adhered soil increased exponentially with the increase of interface temperature; this content firstly decreased and then increased when the moisture content was high. When the soil was dry, there was almost no adhered soil on the surface and the interface temperature had no effect on the adhesive situation. By comparing and analyzing the adhesion state of the soil on the surface of the structure, the influence of temperature on the adhesion characteristics is mainly reflected on the variation of the soil moisture content within the influence range of the interface, the variation of the energy required for the destruction of the adhesion interface, and the change of the location of the weakest antistripping plane induced by both before. This research can better understand the law of formation and development of mud cakes and provided a new idea of solving the problem of mud cakes on the cutter head.

Shield construction in structural soft soils inevitably disturbs the soil around the tunnel leading to tunnel settlement instead of tunnel uplift due to soil unloading rebound during construction. In this paper, tunnel settlement prediction in structural soft soils during construction has been studied by adopting Oujiang Beikou shield tunnel as an example. Firstly, disturbance degree calculation method based on effective stress was introduced and the disturbance degree due to shield construction was numerically simulated. Secondly, combined with one-dimensional consolidation test results, a “two-fold line” settlement calculation model was developed for disturbed soil. Finally, tunnel settlement during construction was predicted through one-way compression layer-summation method and compared with experimentally measured results. Research results demonstrated that the developed method was suitable for the prediction of shield tunnel settlement in structural soft soils.

Accurate prediction of reservoir landslide displacements is crucial for early warning and hazard prevention. Current machine learning (ML) paradigms for predicting landslide displacement demonstrate superior performance, while often relying on various feature engineering techniques, such as decomposing into different temporal lags and feature selection. This study investigates the impact of various feature selection techniques on the performance of ML algorithms for landslide displacement prediction. The Shuping and Baishuihe landslides in China’s Three Gorges Reservoir Area are used to comprehensively benchmark four prevalent ML algorithms. Both static ML models, including backpropagation neural network (BPNN), support vector machine (SVM), and dynamic models, such as long short-term memory (LSTM), and gated recurrent unit (GRU), are included. Each ML model is evaluated under three feature engineering techniques: raw multivariate time series, and feature selection under maximal information coefficient-partial autocorrelation function (MIC-PACF), or grey relational analysis-PACF (GRA-PACF). The results demonstrate that appropriate feature selection methods could significantly improve the performance of static ML models. In contrast, dynamic models effectively leverage inherent capabilities in capturing temporal dynamics within raw multivariate time series, seeing marginal gains with extensive feature engineering compared to no feature selection strategy. The optimal feature selection approach varies based on the ML model and specific landslide, highlighting the importance of case-specific assessments. The findings in this study offer guidance on integrating feature selection techniques with different machine learning models to maximize the robustness and generalizability of data-driven landslide displacement prediction frameworks.

Strain hardening cementitious composite (SHCC) and fiber reinforced mortar (FRM) are considered as suitable for strengthening reinforced concrete (RC) structures, both of which belong to fiber reinforced cementitious composite (FRCC), whereas the distinctive strengthening effects of RC structure using SHCC and FRM have not been sufficiently understood until now. In this study, the comparative study of RC members with flexural strengthening using SHCC and FRM is implemented, focusing on the load carrying capacity and ductility improvement of the strengthened RC member. The result demonstrates that the load carrying capacity of the RC member, even strengthened with a plain SHCC layer, can be improved from that of the RC member with strengthening using a steel-reinforced FRM layer, and the ductility of the RC member with strengthening using a steel-reinforced SHCC layer is significantly improved from that of the RC member with strengthening using plain SHCC layer, because the multiple fine cracks in the SHCC layer are obviously increased even with a low-reinforcement-ratio steel bar adopted in the SHCC layer.