Tunnelling-induced wall damage: An appraisal of elastoplastic constitutive models for masonry

Abstract

This paper investigates the role of masonry elastoplastic constitutive models on tunnelling-induced damage in buildings. A two-stage analysis method (TSAM) is adopted, incorporating input greenfield displacements, 3D masonry walls, and an elastic model for the soil. The paper focuses on four masonry constitutive models that can be readily adopted for routine analysis in industry. Comparison of in-plane yield surfaces with experimental data indicates that, among the considered masonry models, the Concrete Damaged Plasticity model under biaxial calibration gives the best overall performance. The TSAM is then used to study selected tunnel-masonry wall scenarios, confirming a significant effect of the constitutive model and its parameters on masonry wall response to tunnelling, particularly after volume losses where moderate damage is triggered. Also, as masonry stress paths are shown to concentrate in the tensile–compressive areas, with damage prediction being sensitive to the yield surface within this quadrant, numerical damage predictions must rely on the accurate calibration of the constitutive model in the tensile–compressive quadrants. This appraisal indicates that, in the context of routine structure modelling for tunnelling assessments, the selection of elastoplastic masonry models and their biaxial calibration have a non-negligible impact on the damage category estimate.