Scholarlink

<p>Obtaining a velocity model in deep underground engineering is critically important, because both velocity and its variation can reveal the geological structure, excavation damage, and stress evolution. However, the time-lapse seismic exploration for velocity tomography is not practicable for large underground projects. Meanwhile, ill-posed sensor networks in large underground caverns make conventional passive seismic inversion methods hard to converge. To tackle these challenges, this study proposes a new method for three-dimensional (3D) dynamic velocity field inversion based on microseismic (MS) events, named the stratified double-difference (SDD) tomography method. This method solves the challenge of unknown spatial–temporal parameters of MS events by applying double-difference analysis to the observed and calculated travel times. A stratification strategy is implemented to mitigate the adverse effects arising from the spatial distribution of sensors and sources. Based on the realistic MS monitoring set-up of the Jinping (JP) underground laboratory, the provided tomography strategy is tested using two checkerboard models, one containing excavated laboratories, while the other does not. The resolution reaches approximately 50&#xa0;m based on the checkerboard test, and the cross-correlation (CC) between two inverted and true models is 0.84 and 0.79, respectively. Then, dynamic inversion analysis of the excavation process at JP underground laboratory yields a CC value of 0.71. The comprehensive validations demonstrate the ability and reliability of the proposed method for dynamic velocity inversion in the deep-buried large underground cavern.</p>

错误：搜索内容不能为空，请输入英文关键词
错误：关键词超出字数限制，请精简
高级检索

A Stratified Double-Difference Dynamic Tomography of 3D Velocity Field for Deep-Buried Large Underground Cavern

  • 期刊论文

  • 发表日期：

    2026年2月2日

查看全文
  • Yaxun Xiao,
  • Shujie Chen,
  • Liu Liu,
  • Shaojun Li,
  • Guangliang Feng,
  • Jianing Guo,
  • Qiyu Wu

摘要

Obtaining a velocity model in deep underground engineering is critically important, because both velocity and its variation can reveal the geological structure, excavation damage, and stress evolution. However, the time-lapse seismic exploration for velocity tomography is not practicable for large underground projects. Meanwhile, ill-posed sensor networks in large underground caverns make conventional passive seismic inversion methods hard to converge. To tackle these challenges, this study proposes a new method for three-dimensional (3D) dynamic velocity field inversion based on microseismic (MS) events, named the stratified double-difference (SDD) tomography method. This method solves the challenge of unknown spatial–temporal parameters of MS events by applying double-difference analysis to the observed and calculated travel times. A stratification strategy is implemented to mitigate the adverse effects arising from the spatial distribution of sensors and sources. Based on the realistic MS monitoring set-up of the Jinping (JP) underground laboratory, the provided tomography strategy is tested using two checkerboard models, one containing excavated laboratories, while the other does not. The resolution reaches approximately 50 m based on the checkerboard test, and the cross-correlation (CC) between two inverted and true models is 0.84 and 0.79, respectively. Then, dynamic inversion analysis of the excavation process at JP underground laboratory yields a CC value of 0.71. The comprehensive validations demonstrate the ability and reliability of the proposed method for dynamic velocity inversion in the deep-buried large underground cavern.