Upload
john-latilla
View
336
Download
2
Embed Size (px)
Citation preview
Structural and stability assessment for ex-pit dump toe locality optimisation: A case study John Latilla
26 November 2015
BBUGS / BOHOGS Supplier’s day - Moranbah
www.amcconsultants.com
Ukhaa Khudag coal mine, Mongolia
www.amcconsultants.com
Project location
www.amcconsultants.com
Project infrastructure
15 Mtpa CHPP 18 MW Power Station
Office and mess hall Mine Workshops
www.amcconsultants.com
Some key facts
• Multiple seams dipping generally 3° to 17° into highwall
• Flanks (endwalls) dipping 5° to 40° out of pit wall
• Terrace mining (truck and excavator)
• Mid volatile hard coking coal with thermal coal by-product
• Ex-pit waste dumping, at least for next 3 to 5 years
• Nameplate capacity of 15 Mtpa ROM coal to CHPP
• Life-of-Mine stripping ratio of 5.6:1
www.amcconsultants.com
The challenge Optimise the locality of the northern ex-pit waste dump on order to reduce dump truck round trip distances.
Earlier limit equilibrium analyses of cross-sections for pit walls below the dump indicated the potential to steepen the OSA by as much as 4° through buffer blasting the entire slope
Insufficient geotechnical and geological data available to properly identify buffer blasting target zones
The phreatic surface used is based on John Simmons’ work from the Bowen Basin (far wetter conditions)
www.amcconsultants.com
The challenge in a nutshell
7
Possible structural interp. with additional data
LOM structural interp.
New “optimised” crest locality
Potential gain
50m stand-off distance not assessed – remains constant
www.amcconsultants.com
Plan of action
Drill a series of six geotechnical boreholes sited to maximise data from the potential buffer blasted endwall area. Boreholes required to be oriented to enable structural and kinematic analyses
Drill ten percussion holes adjacent to and on top of the waste dump to determine actual phreatic surface
Review geological structure of the ground containing the final endwall slope beneath the waste dump
Conduct kinematic and limit equilibrium analyses to design optimal OSA beneath the waste dump using the updated structural and phreatic surface data
www.amcconsultants.com
Modifications No acoustic televiewer available in country. Changed to drilling oriented cored boreholes
60° inclined core boreholes, oriented by means of Ezy-Mark tool (Reflex). Relies on leaving a crayon impression and a pin profile of the bottom of the hole
Soft and friable coal and mudstone areas - too much fine and loose material at bottom of hole to obtain an impression. Oriented core data unavoidably biased to the stronger units
Inability to orient core in softer zones judged not to be a show stopper. Overall structural orientations expected to be similar in hard and softer rock
www.amcconsultants.com
Modifications continued
Two oriented holes close to the dump toe intersected basement rock (dolerite and basalt) and “replacement” holes were drilled about 100 m further (South) from the dump toe
Info on the basement contacts and structure was nevertheless very useful
www.amcconsultants.com
Oriented core drilling
11
www.amcconsultants.com
Coring through the waste dump
Local exploration drilling contractor used for the dump hydrology holes. Did not have a percussion rig available so drilled the holes with a coring rig
Core was not boxed or logged but simply laid out on the ground near the rig. Recovery of core not called for in the work order
With an air drilled hole the expectation was that water level monitoring would track the gradual climb of water level back up to equilibrium. With water flushed coring the opposite was expected
www.amcconsultants.com
Waste dump core
13
www.amcconsultants.com
Structural data collection
Oriented core holes logged and photographed by the site geotechnical team to CoreLog standard (ACARP 2015)
Individual defects identified and measured using CoreProfiler (CSIRO). Builds a composite photograph of the core column from core photographed in trays. From this, dip and dip direction are determined for each defect and plotted on stereonet. Data was then exported to Dips6.0 for further manipulation and assessment
CoreProfiler also produces RQD and fracture frequency data
CoreProfiler is still in development but was found to be a useful tool
www.amcconsultants.com
CoreProfiler – core box photo From this:
15
www.amcconsultants.com
CoreProfiler work page example To this:
16
www.amcconsultants.com
Structural assessment
Structural data analysed in Dips6.0 as follows:
• All defects in all boreholes, then split per endwall sector (NEW2 and NEW4) and finally split into individual boreholes
• Each of the above were then sub-divided into:
All structure above the lowermost coal seam (BOC) and that below BOC – the latter being the key strata for endwall design
Bedding planes only split as above to determine structural trends in sedimentary rock
• Shears and possible shears analysed separately – insufficient numbers to analyse shears on a hole for hole basis.
www.amcconsultants.com
Selected stereonet plots
18
All shears – all b/hs
All defects – all b/hs
www.amcconsultants.com
Structural analysis summary
Description Dip and Dip direction
(°) Set 1 Set 2 Set 3
All defects for all boreholes 13/207 50/174 – All defects above BOC 11/211 50/167 – All defects below BOC 13/204 49/180 – Bedding planes only for all boreholes 13/208 49/179 –
Bedding planes only above BOC 14/213 52/173 – Bedding planes only below BOC 12/203 47/183 – All defects for basement rocks 11/214 51/180 – Shears only for all boreholes (orientation low to medium confidence)
64/172 73/193 23/217
www.amcconsultants.com
Kinematic assessment
Critical Surface Limits
• Batters
Final endwalls – 10%
Interim endwalls – 15%
Highwalls – 20%
• Overall slope angles
Below critical infrastructure – 5%
All others – 10%
www.amcconsultants.com
Kinematic assessment
Planar and wedge sliding are the most likely, kinematically admissible, failure types
• Excluding the effect of shears:
Batters should not be formed steeper than 55° in NEW4 and 70° in NEW2
Overall slope angle (OSA) should not exceed 47°
• Taking shears into account:
Batter slope angle drops to 25°. OSA 23° for NEW2 and 30° for NEW4
• Assuming bedding shears will be buffer blasted:
Max batter angle 55° and OSA 27° to 34°
www.amcconsultants.com
Selected stereonet plots
22
NEW2 batter scale planar sliding
NEW2 batter scale wedge sliding
www.amcconsultants.com
Kinematic analysis results summary
Slope scale Pit sector
Maximum slope dip angle and percentage of critical surfaces
Planar sliding (° / %)
Wedge sliding (CS Mod)*
(° / %)
Batters (15% critical surfaces limit)
NEW2 70/13.7 70/13.2*
NEW4 55/14.9 67/15.0*
NEW2 shears 28/14.3 25/13.2
NEW4 shears 59/7.1 33/13.2
OSA (10% critical surfaces limit)
NEW2 49/9.8 47/9.8*
NEW4 47/9.7 49/9.8*
NEW2 shears 27/7.1 23/9.9
NEW4 shears 59/7.1 30/6.6
*The critical surface results for wedge sliding have a correction factor of 0.5 applied to better match site experience. This is referred to as the modified wedge sliding critical surface value (CS Mod.).
www.amcconsultants.com
Structural interpretation
General trend indicated steeper than originally modelled coal bearing strata along the northern boundary fault in 3 out of 4 cross-sections (CS 20a, 22 & 23)
The other cross-section (CS 28) indicated steeply dipping boundary fault cutting off the coal strata
A stronger and less disturbed conglomerate zone was identified below the BOC
Zone of steeply dipping sedimentary rocks assumed to extend along the southern face of the northern boundary fault. Based on previously exposed fault zones at UHG. Width variable
www.amcconsultants.com
Dump toe and borehole locality
25
Cross section lines
Geotech boreholes Arrow denotes hole azimuth
Hydrology boreholes
www.amcconsultants.com
Structural interp. CS 20a
26
www.amcconsultants.com
Phreatic surface model
Locality
Distance from toe (m)
Phreatic surface below original ground surface (m)
Minimum Maximum Minimum Maximum Generic model
240 m from dump toe (on pit side) – 240 – – 28
~100m from dump toe 50 127 18 29 23
Immediately below dump toe 0 0 13 14 13
~60 from toe under dump 40 92 3 6 3
~180 from toe under dump 151 230 0 0 0
210 m from toe under dump 200 230 – – +2
[1] Note that where the pit crest is closer than 240 m the generic (or specific if known) phreatic surface model must be merged with this model
www.amcconsultants.com
Phreatic surface interp. CS 28
28
www.amcconsultants.com
Limit equilibrium analyses
Based on oriented borehole data and observations of core recovered from hydrological drilling the interpreted structure from the 2014 model has been modified
The endwall crest can be located up to 180 m further south
Existing minimum stand-off distance between dump toe and slope crest of 50m was not investigated or altered in this study
The endwall areas represented by the two eastern most cross sections (CS 23 & 28) require no additional buttressing or dewatering
The most westerly, and deepest, endwall slope (as represented by the two most westerly cross sections, CS 20a & 22) requires extensive toe stabilisation plus dewatering
www.amcconsultants.com
Cross-section 20a
30
Buffer blast
Waste buttress
www.amcconsultants.com
Cross-section 28
31
G A L EN A Version 6.10
Project UHG 2014 Slope Stabi l i ty A naly sis - Cross Sect ion 28
Fi le: C:\Users\Jlati l la\Documents\Numerical model l ing\Galena\UHG\2015 N EW4 Buf fered endw al l below dump CS 28 (extended).gmf
Edited:Processed:
21 A ug 201521 A ug 2015A MC Consul tantsL icensed to:
A nalysis
Resul ts
4Mul t iple Stabi l i ty A nalysisMethod:Surface:
Spencer-WrightCircular
Cri tical Factor of Safety : 1.24I ntersl ice Force (Final ) A ngle: 28.9 °
Material K eysW OB (c:27 ø:22.4 g:22)
MWOB (c:115 ø:31.7 g:23)
FOB (c:358 ø:37.2 g:24.5)
CF (c:40 ø:23 g:14.4)
BPS (c:0 ø:14 g:19.4)
W R (c:27 ø:26.8 g:18.3)
FSZ (c:24 ø:18 g:21.8)
CG (c:785 ø:42.9 g:24.7)
BW (c:612 ø:34.1 g:29)
BF (c:828 ø:41.1 g:29)
800 900 1000 1100 1200 1300
1200
1250
1300
1350
1400
1450
1500
1550
1600
1650
Buttress of flat to shallow dipping strata
www.amcconsultants.com
Conclusions
This extensive study has resulted in far better understanding of the structure and strength of critical strata beneath the waste dump toe
This improved knowledge has enabled the dump toe locality to be moved significantly closer to the pit crest thereby significantly reducing dump truck round trip distances
Thank you
Energy Resources LLC are thanked for their assistance in preparing this presentation and for permission to share this experience