Multi-Scale Analysis of Tunnelling-Induced Settlement Effects on Reinforced Concrete Frames: Integrating Numerical, AI, and Froude-Similitude Experiments

The rapid urbanization and increasing demand for underground transportation systems have led to significant advancements in tunnelling technologies. However, tunneling-induced ground settlements pose critical challenges to the safety and integrity of existing structures, particularly in densely populated urban environments. This thesis addresses these challenges by integrating geotechnical and structural engineering perspectives to comprehensively assess tunnelling effects on structures, offering novel solutions to bridge existing gaps in research and practice. A novel artificial intelligence-driven model has been developed to accurately predict tunnel-induced settlements in cohesionless soils, offering a reliable and practical tool for early-stage design and planning. A simplified modelling approach has been introduced to efficiently evaluate tunnelling-induced surface settlements and their effects on adjacent structures, striking a balance between computational efficiency and predictive precision. Advanced numerical simulations delve deeper into the structural response of reinforced concrete (RC) frames to differential settlements caused by tunnelling, incorporating complex soil-structure interactions and nonlinear structural behaviour for a more comprehensive analysis. Innovative scaled physical models were developed to address the scarcity of experimental data on settlement effects. These experiments provide critical insights into the response of RC structures under differential settlements while pioneering a more efficient physical modelling technique. The study successfully replicates real-world structural behaviour by employing novel material scaling methods and similitude principles, bridging the gap between theoretical predictions and practical applications in structural design and resilience. The thesis further extends its scope by examining the interplay between differential settlements and seismic forces, providing critical insights into the combined vulnerabilities of RC frames under multi-hazard scenarios. The findings contribute to refining numerical models, enhancing predictive accuracy, and informing resilient structural design practices. By addressing both theoretical and practical aspects, this research offers a robust framework for mitigating the impacts of tunnelling on existing structures and advancing the resilience of urban infrastructure amidst complex engineering challenges.