Actual seismic events typically exhibit mainshock-aftershock sequence characteristics, and source characteristics have a significant impact on cavern response. Currently, the influence of near-fault mainshock-aftershock sequences (NFMA) and far-field mainshock-aftershock sequences (FFMA) on underground caverns is generally ignored. This study aims to establish a framework for evaluating the dynamic response characteristics and seismic fragility of large-scale underground caverns under NFMA/FFMA. The response laws of residual displacement and rock fracture degree of cavern under NFMA/FFMA are comparatively studied, and the failure probability of different damage states is quantified by the fragility function. The results show that the surrounding rock of underground caverns exhibits significant cumulative damage effects and non-uniform failure characteristics under mainshock-aftershock sequences. Aftershock fragility is strongly related to the mainshock-damaged state for underground caverns. The collapse probability of underground caverns after 0.9g aftershocks in NFMA increased from 0.76% in slight damage to 21.12% in moderate damage and 53.51% in severe damage. This study can provide a probabilistic basis for seismic design, aftershock risk warning, and post-earthquake emergency assessment in underground engineering.

Hyperspectral imaging provides a novel approach for intelligent geological perception in tunnelling and underground engineering due to its high spectral resolution, nondestructive nature, and combined spectral-spatial information. However, in confined underground spaces, noise is often introduced by short exposure times, low illumination, and dust, and limited spatial resolution can cause mixed pixel effects, complicating data processing. This study presents an underground hyperspectral imaging-based mineral mapping method that achieves wall-rock visualization and semi-quantitative mineral mapping through image denoising and spectral unmixing. A spatial-spectral recurrent transformer U-Net is developed to reduce noise by leveraging spectral band correlations and nonlocal spatial-texture dependencies. A Dirichlet-based mixed pixel simulation is used to address spectral mixing, with the N-FINDR algorithm identifying endmember minerals, and the fully constrained least squares method to estimate mineral abundances. When applied to a water diversion tunnel in Shanxi, the method generates spatial distribution maps of dolomite and calcite. The experimental results confirm its effectiveness for intelligent geological logging and subsurface geological feature analysis.