Author ORCID Identifier

https://orcid.org/0000-0001-8942-3314

Semester

Spring

Date of Graduation

2025

Document Type

Dissertation (Campus Access)

Degree Type

PhD

College

Statler College of Engineering and Mineral Resources

Department

Mining Engineering

Committee Chair

Deniz Tuncay

Committee Co-Chair

Thomas M. Barczak

Committee Member

Thomas M. Barczak

Committee Member

Hung-Liang (Roger) Chen

Committee Member

Qingqing Huang

Committee Member

Deniz Talan

Abstract

In the United States, most underground coal mine roofs consist of shale, often referred to as shale roofs. The presence of bedding planes makes shale exhibit complex anisotropic and brittle failure behaviors, especially under mining-induced stress, making these roofs highly susceptible to collapse. Traditional geomechanical models often oversimplify bedding planes as continuous and equidistant. This dissertation applies image processing to capture realistic bedding plane characteristics and integrates numerical simulations to study brittle failure characteristics and reinforcement strategies for shale roofs. The research spans multiple scales, from laboratory specimens to entry-scale mine models, culminating in the development of a ground reaction curve (GRC) methodology for optimizing standing support design.

Firstly, an image processing method was applied to capture bedding plane characteristics across different orientations, scales, and shale types. The extracted bedding plane data—coordinates, number, spacing, and length—were compiled into a comprehensive database. Secondly, using the image-processing-based bedding plane extraction, a laboratory-scale numerical model was developed to simulate the brittle failure of shale specimens. This study developed a comprehensive calibration procedure that incorporated bedding plane anisotropy into numerical models. The procedure involves component analysis of shale, capture of bedding planes through image processing, initial micro-property selection, assessment of anisotropic effects, and validation of macro-properties against laboratory data. The results demonstrated that the calibrated UDEC model successfully replicated stress-strain behavior, failure modes, and anisotropic effects across bedding plane orientations. These models provided detailed insights into shale’s anisotropic brittle failure characteristics, showing that strength, Young’s modulus, and failure modes depend on bedding plane orientations.

Thirdly, extending to the entry scale, the anisotropic brittle failure characteristics of shale roofs were investigated across five U.S. coal seams: Lower Kittanning, Pittsburgh, Pocahontas No. 3, Blue Creek, and Sunnyside. Vertical stress values ranged from 5 MPa to 48 MPa, while horizontal stresses varied from 6 MPa to 42 MPa, covering diverse roof geological conditions. An anisotropic brittle failure criterion was developed and incorporated into a FLAC3D entry-scale model to capture the strength and Young’s modulus anisotropy due to bedding plane orientations obtained in the laboratory. The developed anisotropic brittle failure criterion was verified in a single-element model, multiple elements model, and entry-scale model. Particularly, the entry-scale model was calibrated by underground measurements of stress path, roof sag, and cable loads. Finally, based on the calibrated entry-scale model, a numerical methodology was developed to generate local geology- and stress-dependent GRCs. Seven case studies from the aforementioned five coal seams were used to calibrate and validate the methodology.

By integrating image processing, lab-scale models, entry-scale model analysis, and standing support optimization, this dissertation presents a comprehensive approach to understanding and mitigating shale roof failures in underground coal mines where standing support is utilized.

Download

Share

COinS