Nonlinear seismic analysis of subway station − layered site under obliquely incident shear waves with arbitrary angles

Abstract

Accurately capturing the seismic response of underground structures subjected to obliquely incident seismic waves, particularly when the angle of incidence surpasses the critical value, remains a challenging task in earthquake engineering. To address this gap, this paper presents a three-dimensional (3D) nonlinear seismic analysis of subway stations embedded in a layered site, specifically in response to obliquely incident shear (SV) waves at arbitrary angles. An innovative procedure, termed the coupled dynamic stiffness matrix–finite element method (DSM-FEM), is introduced to enable seismic input by transforming responses induced by arbitrarily incoming SV waves into equivalent nodal loads. To accurately simulate wave propagation within the site, a viscous-spring artificial boundary is utilized, while a nonlinear generalized Masing model that incorporates modified damping is employed. Using the Daikai subway station as a benchmark, the research examines the effects of varying oblique incident angles on the structural response, taking into account dynamic soil-structure interaction. The results reveal that the maximum response, including peak deformation, internal forces, Mises stress, occurs when the incident angle approaches the critical value. Beyond this critical angle, the seismic response notably diminishes. Additionally, the influence of horizontal incident angles is found to be noticeable, leading to variations in deformation patterns and internal forces across different structural components. Specifically, it has been observed that the drift ratio, displacement, shear force, acceleration, and Mises stress exhibit a decreasing trend as the horizontal incident angles increase. These findings highlight the significance of considering non-vertical input ground motion in seismic analysis, and offer valuable insights for the structural design and safety evaluation of underground structures.