Analytical Solutions and Parameter Sensitivity of Construction Disturbance–Induced Settlement of a Prefabricated Underground Station

Résumé fourni par la source

During the construction of prefabricated underground stations (PUSs), adverse geological conditions and external loads can cause deformation and misalignment of the station rings. Previous models treat PUS as a continuous beam, neglecting the discontinuous deformation between joints. This study formulates a finite-difference equation for the beam–spring model on a Pasternak elastic foundation, accounting for the effects of circumferential joint rotation and displacement misalignment. For the issue of longitudinal settlement deformation in PUS structures, a simplified solution was derived and a computational model was established to analyze deformation behavior under arbitrary upper load conditions when the boundary conditions of prefabricated assembly segments are rigid connections. The sensitivity of the settlement model to variations in key parameters was also discussed. A machine-vision monitoring system was deployed on-site to monitor the settlement deformation of segment rings in the PUS. The accuracy and applicability of the simplified solution were verified by comparing the computational model results with measured data. The research findings indicate that (1) after the completion of the backfill, the measured maximum longitudinal displacement is 4.58 mm, slightly lower than the theoretical value of 4.87 mm, which confirms the reliability of the established model; (2) variation in the foundation reaction coefficient, rotational stiffness of circumferential joints, and shear stiffness of circumferential joints significantly influence the settlement deformation of the station structure; when these three parameters are each set to 0.5 times their original values, the maximum structural settlement increases by 28.74%, 31.01%, and 8.62%, respectively; and (3) the machine-vision monitoring system recorded a maximum displacement deviation of 0.36 mm over a distance of 70 m, with the overall average error not exceeding 5%, thereby demonstrating high accuracy in structural displacement monitoring within a defined range.

Ce résumé expose les affirmations des auteurs. BNTIC ne l’interprète pas comme une validation indépendante des résultats.