| |
|
| Earth and Rock Works | Earth Retaining Structures | Geotechnical Hazards | Ground Improvement | Structural Foundations | Subsurface Investigation |
| FHWA > Engineering > Geotechnical > Geotechnical Hazards > Mines |
| Subject | From | Posted |
|---|---|---|
| Sunsidence of room and pillar coal mines | Greg Hawkins | 11/30/2003 08:29PM |
| RE: Sunsidence of room and pillar coal mines | Matt DeMarco | 01/20/2004 11:28AM |
| RE: Sunsidence of room and pillar coal mines | Marc Bétournay | 01/21/2004 09:23AM |
Messages posted for Sunsidence of room and pillar coal mines
|
Subject: Sunsidence of room and pillar coal mines From: Greg Hawkins Date: 11/30/2003 | ||
| Does anyone have any suggestions on the best way to estimate settlement and surface strain, tilt curvature folling the collapse of bord and pillar workings at 80 m to 300 m depth? Thanks Greg |
||
|
Subject: RE: Sunsidence of room and pillar coal mines From: Matt DeMarco Date: 01/20/2004 | ||
| Greg-- The setting you are describing can be complicated for several reasons. Bord-and-pillar (called room-and-pillar in the states) mining systems may not have sufficient panel widths to effect supercritical subsidence (full flat-bottomed trough subsidence) at the higher end depth range your asking about, which complicates analyses. Also, the nature of the panel mining can make for non-uniform subsidence due to partial pillar failures, existence of retreat mining areas, split pillar mining methods, and unknown impacts of topography (if mountainous). A big problem is estimating what portion of the entire workings has actually collapsed, and to what degree. We usually can't go in and inspect the extent of the failures, so we need to make some judgements as to what we expect. The best approach I have ever used involves a combination of plan-view, psuedo-3D boundary element modeling and conventional influence function empirical modeling. The boundary element modeling is used to determine the actual failure progression of the workings and the vertical surface strains and deformations over failed areas (Mohr-Coulomb failure analysis). MULSIM/NL, available from Keith Heasley at West Virginia University, is an ideal BE model especially suited to this problem. The outputs show the extent of pillar failure for the actual mine layout, and gives vertical strain profiles at the surface. Once the surface vertical deformations are determined, curve fitting to an influence function method is required to then extract infomation on horizontal strains, curvature, tilt, inflection locations, etc. MULSIM/NL will not give information on horizontal surface strains, but provides very good matches to those vertical strains measured in the field. For supercritical panels - usually having a minimum failed width of 1.4 times the depth - total subsidence typically falls from 50-70% of the extracted seam height (single seam). MULSIM/NL will give similar results, and is typically within 10-25% of the total vertical strain actually measured. I would call Tom Vandergrift at NSA Engineering, Golden, CO, for more details on how this combined method works. His number is 303-277-9920. NSA has a proprietary topography mesh generator that really speeds up the analysis. Tom can send you some published papers on the subject. |
||
|
Subject: RE: Sunsidence of room and pillar coal mines From: Marc Bétournay Date: 01/21/2004 | ||
| Greg, I think Matt has probably well outlined the trough subsidence resultant and best approach (numerical modelling) that I would use as well. I dont think that sinkholes (chimneying to surface) caused by localized roof collapse will be an issue at those depths, the caving probably choking itself off. In any case, single chimneying mobilizes very little of the surrounding rock mass on its path towards surface, so that settlement normally occurs only when the failure front reaches the overburden. In rare cases, multiple chimneying areas can be initiated. This can lead to leading to loss of confinement over pillars and leads to general rock mass failure and resultant subsidence trough. Again numerical modlling would be your best tool. |
||
We are sorry this forum is currently closed. If you have a topic you would like to add, pleas contact Tom Lefchik
