highwall stability due to ground vibrations from blasting dr. kyle a. perry dr. kot f. unrug kevin...

25
HIGHWALL STABILITY DUE TO GROUND VIBRATIONS FROM BLASTING Dr. Kyle A. Perry Dr. Kot F. Unrug Kevin Harris Michael Raffaldi

Upload: arthur-houston

Post on 14-Dec-2015

216 views

Category:

Documents


2 download

TRANSCRIPT

HIGHWALL STABILITY DUE TO GROUND VIBRATIONS

FROM BLASTING

Dr. Kyle A. Perry

Dr. Kot F. Unrug

Kevin Harris

Michael Raffaldi

PURPOSEObjective: Study the effect of blast vibrations

on the stability of highwalls and web and barrier pillars

Methods• Numerical Modeling• Field Testing• Laser Scanning• Vibrations Monitoring

PROGRESS

SASW Testing

Numerical Modeling

• FLAC3D – Web and Barrier Pillars– Stress Distribution– Pillar Stability

• 3DEC – Highwall Modeling– Laser Scanning– Displacement/Velocity

Field Data Collection

SASW Testing

Dynamic Material Properties

• Shear Modulus

• Damping Ratio

Modulus/Damping Reduction Curves

• Dynamic Model Calibration

• Iterative Numerical Process

SASW Testing4000

3000

2000

1000

0

Su

rfa

ce W

ave

Ve

loci

ty (

ft/s

)

3 4 5 6 7 8 910

2 3 4 5 6 7 8 9100

2 3

Wavelength (ft)

Experimental Theoretical

Numerical ModelingSub-Objective 1: Evaluate the influence of

highwall mining progression on web/barrier pillar stability using FLAC3D

Methodology• Simplification – 2D Section• Validation – Mark-Bieniawski• 5 Case Studies

Material PropertiesCoal Material Model • Strain Softening Mohr-Coulomb

Roof/Floor Interface• Bi-linear Mohr-Coulomb

Property Value

Elastic Modulus (GPa) 3

Poisson Ratio 0.25

Cohesion (MPa)  

Initial 1.81

Final 0.41

Friction Angle (deg)  28

Plastic Strain Range 0.015 0 1.15 2.3 3.45 4.6 5.75 6.90

0.5

1

1.5

2

2.5

Normal Stress (MPa)

Sh

ear

Str

ess

(MP

a)

Material PropertiesRock Material Model• Ubiquitous Joint Model with Softening

Rock Matrix Properties Bedding Plane Properties UCS Elastic 

ModulusPoisson Ratio

Friction Angle Cohesion Tensile 

Strength Dilation Friction Angle Cohesion Tensile 

Strength MPa GPa --- Deg MPa MPa Deg deg MPa MPaSS 100 40 0.25 40 13.52 5.8 12 30 6.76 4.64SH 30 10 0.25 20 7.3 1.74 19 7 0.5 0.17

Matrix SofteningCohesion Tensile Strength Dilation

Maximum Residual Range* Maximum Residual Range Maximum Residual Range*MPa MPa --- MPa MPa --- Deg Deg ---

SS 13.52 0 0.005 5.8 0 0.001 12 0 0.005SH 7.3 0 0.005 1.74 0 0.001 19 0 0.005

Bedding Plane SofteningCohesion Tensile Strength Dilation

Maximum Residual Range* Maximum Residual Range Maximum Residual Range*MPa MPa --- MPa MPa --- Deg Deg ---

SS 6.76 0.68 0.005 4.64 0 0.001 12 0 0.005SH 0.5 0.05 1.005 0.17 0 0.001 19 0 0.005

Model Validation

1 2 3 4 5 6 7 8 9 10 110

5

10

15

20

25

30

Mark-Bieniawski Model

W/H Ratio

Pea

k S

tren

gth

(M

Pa)

0 1 2 3 4 5 6 70

10

20

30

40

50

60

70

80

Pillar 1 Pillar 2 Pillar 3 Pillar 4

Distance Into Rib (m)

Str

ess

(MP

a)

Model Validation

0 0.2 0.4 0.6 0.8 110

11

12

13

14

15

16

17

18

W/L Ratio

Pea

k S

tren

gth

(M

Pa)

2.5 3 3.5 4 4.5 5 5.5 6 6.5 710

12

14

16

18

20

22

24

26

28

Mark-Bieniawski Model (Elastic)Model (SS) Model(SH)

W/H Ratio

Pea

k S

tren

gth

(M

Pa)

Model Validation

Material Properties

• Peak Strength

• Calibration Reasonable

Web/Barrier Pillars

• “Infinite Length” Assumption

2D Cross Section

• Plane Strain Validation

Case Study ResultsCase Height (m) Web (W/H) Barrier (W/H) Cover (m)

C1 1.219 0.75 3 45.7

C2 1.219 1 4 60.9

C3 1.219 1.25 4 70.1

C4 1.219 1.5 5 91.4

C5 1.219 1.75 6.5 118.9

CS1 CS2 CS3 CS4 CS50.00

1.00

2.00

3.00

4.00

5.00

6.00

7.00

8.00

SSTONE

ELASTIC

SHALE

TRIB. AREA

Ave

rage

Web

Str

ess

(MP

a)

CS1 CS2 CS3 CS4 CS50.00

1.00

2.00

3.00

4.00

5.00

6.00

7.00

8.00

SSTONE

ELASTIC

SHALE

TRIB. AREA

Ave

rage

Bar

rier

Str

ess

(MP

a)

Mining Progression Results

Web 1

Web 2

Web 3

Web 4

Web 5

Web 6

Web 7

Web 8

Web 9

Web 10

80%

82%

84%

86%

88%

90%

92%

94%

96%

98%

100%

CS1

CS2

CS3

CS4

CS5

Important Findings

Tributary Area Loading

• Common Assumption

• Arching Effect – Geology Dependent

• Importance of Barrier Pillar Design

Mining Progression

• De-stressing near Barriers

• Asymmetric Loading

Probability of Pillar Failure/Highwall Collapse

Dynamic Impact on Web/Barrier Pillars

Numerical ModelingSub-Objective 2: Correlate active highwall

sites with 3DEC for validation and additional testing

Methodology• Routine trips collect scan data• Digitize geometry for mesh generation

within 3DEC

Primary Mine SitePine Branch Mine• Perry County, KY• Mountaintop/Contour• Hazard 7 – Hazard 10 seams• 200’ – 300’ Highwall• Massive Sandstone

Important Mine Data• Corehole Data• Mapping

Primary Mine Site

VIBRATION AND SCAN DATA

Laser Scanning• Pre/Post Blast Capture• Mesh Generation for 3DEC• Geologic Data

– 2 Primary Joint Sets– 85 Degree Dip– 110, 185 Direction

Seismograph Deployment• Highwall Peak and Bottom• Single Event Vibration Data

SUMMARYHIGHWALL

STABILITY STUDY

MODELING FIELD DATA

LIDARVIBRATION RECORDS

DESIGN CALIBRATION

FIELD TESTING

LITERATURE/ EMPIRICAL

COMPLETED MODEL

BLAST AND DAMAGE CORRELATION

NUMERICAL STUDY

DRAW CONCLUSIONS

Project DirectionFLAC3D Modeling• Probability of Pillar Failure/Highwall Collapse

• Dynamic Impact on Web/Barrier Pillars

3DEC Modeling

• Solidify Static Model

• Quantify/Correlate Damage and Vibrations

• Sensitivity Studies

– Ratio of PPV

– Geology

– Highwall Geometry

QUESTIONS?