Investigation of Thermo-Hydro-Mechanical Behaviour of Energy Tunnels Using Advanced Soil Constitutive Models

The transition to low-carbon energy systems requires scalable and sustainable solutions for heating and cooling, which together account for nearly half of global energy demand and a substantial proportion of greenhouse gas emissions. Shallow geothermal systems have emerged as an important renewable option, particularly in urban areas where the subsurface can be harnessed to provide low-carbon thermal energy. Among these, energy tunnels, transport tunnels equipped with heat-exchange pipes embedded in their linings, represent one of the most promising geo-energy technologies. By integrating thermal energy harvesting with essential infrastructure, energy tunnels can contribute to both sustainable urban development and the decarbonisation of heating and cooling. Despite their potential, the understanding of energy tunnel behaviour remains limited. Most studies to date have concentrated on thermal efficiency, while relatively few have investigated their mechanical response. Those that do typically rely on simplified constitutive soil models such as linear elasticity or Mohr--Coulomb plasticity. These formulations cannot capture key mechanisms such as stiffness degradation at small strains or thermoplastic soil behaviour, and therefore exhibit substantial limitations for modelling the coupled thermo-hydro-mechanical (THM) behaviour of energy tunnels. As a result, existing approaches risk misrepresenting ground deformations, pore pressure evolution, and lining stresses, particularly under long-term cyclic thermal loading in fine-grained soils. In addition, the interaction between multiple tunnels operating in close proximity has not been systematically investigated, despite its growing relevance in densely built environments. These limitations constrain the establishment of reliable design frameworks and may compromise both structural performance and long-term energy efficiency. This thesis addresses these challenges by advancing the analysis of energy tunnels within a fully coupled THM framework. The research is structured around three major objectives. The first is the adoption of an elasto-plastic critical state soil model, enhanced through the integration of small-strain stiffness, in order to improve the prediction of energy tunnelling-induced deformations and cyclic thermal response. The second is the investigation of thermal interaction in twin energy tunnels, where the coupled THM response is analysed under different operational scenarios, with particular attention to the influence of soil permeability and multiphysical couplings. The third is the development of a thermo-plastic extension of the Modified Cam-Clay model, allowing the yield surface to evolve with both plastic volumetric strain and temperature, and enabling the simulation of irreversible volumetric strains such as thermal collapse and hardening, as well as long-term consolidation effects. The findings demonstrate that the choice of constitutive model plays a decisive role in the analysis of energy tunnels. Incorporating small-strain stiffness within an elasto-plastic framework was shown to significantly improve predictions of deformations and soil-structure interaction, whereas simplified elastic approaches underestimated settlements and pore pressure response. Analyses of twin tunnels revealed that thermal interaction depends strongly on operational strategy and soil permeability, and identified two distinct settlement mechanisms: temperature-induced contraction of the soil skeleton and consolidation driven by negative excess pore pressures, the latter captured only in fully coupled THM simulations. This dual mechanism explains the permeability-dependent response and underscores the importance of accounting for hydro–mechanical coupling when assessing long-term tunnel performance. Finally, the thermo-plastic extension of the Modified Cam-Clay model successfully reproduced irreversible volumetric strains observed in laboratory tests, capturing both thermal collapse in normally consolidated states and elastic expansion in overconsolidated clays. Applied to energy tunnels, this model highlighted the importance of accounting for thermal loading history demonstrating the inadequacy of conventional linear elastic thermal expansion approaches. Beyond their theoretical contributions, these results have practical implications for geotechnical and energy-geostructure engineering. They enable more reliable predictions of settlements, pore pressure evolution, and lining stresses; they emphasise the need to explicitly consider multi-tunnel interaction in urban design; and they provide a foundation for incorporating thermoplasticity into design frameworks. Collectively, this work advances the scientific understanding of energy tunnels and contributes to establishing the basis for their safe, efficient, and sustainable implementation in cities.