Abstract
The underground powerhouse of a hydropower station in Northwest China features a large height-to-span ratio and complex geological conditions, posing critical challenges for surrounding rock stability. This study employs a systematic framework of “parameter inversion–stability analysis–support optimization”. The main contributions are: (1) A forward-backward analysis method integrating monitoring data, numerical calculation, and parameter optimization was used to invert the physical and mechanical parameters of the surrounding rock, establishing a practice-consistent numerical model. (2) Considering complex geological conditions from a prior collapse, systematic analyses of deformation characteristics, stress distribution, and overall stability during subsequent bench excavation were conducted, revealing evolutionary patterns of deformation, reproducing the collapse mechanism, and evaluating rock mass safety. (3) An optimized reinforcement scheme was proposed, involving additional anchor cables at arch shoulders and two anchor bolts on each sidewall (Sidewalls 2), with comparative analysis confirming its feasibility. Key findings: (i) Based on the inversion of monitoring data, the obtained physical and mechanical parameters of the surrounding rock can effectively reflect the actual characteristics of the rock mass and align with monitoring trends. (ii) Stability analysis of subsequent layered excavation, accounting for prior collapse impacts, shows that while deformation at key points continued to increase, it remained below critical thresholds, maintaining overall stability—prestressed anchor cables significantly controlled deformation. (iii) The optimized support scheme yields comparatively favorable outcomes with respect to the containment of surrounding rock displacement and the internal forces within the supporting. These outcomes provide valuable references for the design and construction of similar underground powerhouses.
Funding Information
  • Scientific Research Program of Shaanxi Provincial Education Department (24JK0389)

This publication has 19 references indexed in Scilit: