Energy Release and Failure Characteristics of Coal Samples: Laboratory Test and Numerical Modelling

Ting Ren, Xiaohan Yang and Lihai Tan

4th International Symposium on Dynamic Hazards in Underground coalmines, 22-23 June 2019, CUMT, Xuzhou China

University of Wollongong, Australia 澳大利亚伍伦贡大学

QS Ranking 2020212

3

Coal Burst in Australian U/G Coal Mines

4

• Coal ‘Bumps’ have also been reported by other mines in NSW

• Awareness of potential Burst hazards in QLD

Coal Burst in Australian U/G Coal Mines

Structural Geology of Coal Burst Sites

Coal Burst in Australian U/G Coal Mines

Energy AnalysisStatic and Dynamic Load Superposition TheoryCoal burst will occur when the sum of static and dynamic load exceeds the minimum load required for coal burst formation. The energy released during coal burst is provided by static load and dynamic load.

Coal Burst Induced by Static and Dynamic Load superposition (Dou et al)

Energy AnalysisEnergy Sources of Coal Bursts in AustraliaElastic energy accumulation resulted from high mining depth and complicated geological structure is the major contribution of energy sources of coal burst.

01234

2014/3/2 2014/9/18 2015/4/6 2015/10/23 2016/5/10 2016/11/26 2017/6/14 2017/12/31 2018/7/19

Mag

nitu

de(M

L)

Date

EarthquakeCoal Bursts

01234

2014/6/10 2014/12/27 2015/7/15 2016/1/31 2016/8/18 2017/3/6 2017/9/22 2018/4/10 2018/10/27

Mag

nitu

de(M

L)

Date

EarthquakeCoal Bursts

Coal Burst of Coal Mine A

Coal Burst of Coal Mine B

8

WET is the indicator of the proportion of elastic energy storage of coal when coal is near critical stress.

Coal samples with low KE value will fail gentler as more energy is dissipated by deformation.

The violence of coal burst reflects in the instantaneous of energy releasing as well (WB Zhang et. al, 1986).

According to our analysis, elastic energy storage of coal samples increases with uniaxial compressive strength (UCS) ranges from 0 to 50.

Elastic strain energy index (WET)

Bursting energy index (KE)

Dynamic failure time (DT)

Uniaxial compressive strength (RC)

Coal Burst Propensity Index

Coal Burst Propensity Index

9

Sample Preparation1

Sample Measuring2

RC and KE Test3 Calculation4 WET Test5 DT Test6

Loading Machine and Control SystemRadial Coring Drill Machine Coal Sample with Strain

Gauges

Risk Classification7

Coal Burst Propensity Index

10

Load/kN

0

5

10

15

20

25

Time/s0 10 20 30 40 50 60 70 80 90 100 110 120

DT Test Curve

Str

ess

/M

pa

0

2

4

6

8

10

12

Time/ s0 20 40 60 80 100 120 140 160 180 200

RC Test Curve

Str

ess

/M

pa

0

2

4

6

8

10

12

Strain/με0 1,000 2,000 3,000 4,000 5,000 6,000 7,000 8,000 9,000 10,000

KE Test Curve

Str

ess

/M

pa

0

2

4

6

8

10

12

14

16

18

20

Strain/με0 1,000 2,000 3,000 4,000 5,000

WET Test Curve

Intact Sample Failed Sample Intact Sample Failed Sample

Intact Sample Failed Sample Intact Sample Failed Sample

Coal Burst Propensity Index

Coal Burst Propensity Index

12

Quantitative Study of Coal Burst Energy

13

Energy Accumulation and Releasing of Coal Burst

Quantitative Study of Coal Burst Energy

The relationship between elastic energy and plastic energy of coal samples can be measured by coal burst propensity index. The relationship between the various energy forms of coal samples, in particular the relationship between elastic energy and burst energy, acoustic emission energy and burst energy will need future research.

Energy Analysis

Schematic Diagram of Energy Accumulation before Peak Strength

Stress versus Strain Curve of Coal Samples

Load/kN

0

5

10

15

20

25

Time/s0 10 20 30 40 50 60 70 80 90 100 110 120

Fitting Functions of Fragment Size DistributionCoal Ejection Test

Energy AnalysisKinetic Energy EstimationThe estimated kinetic energy by ejected coal is between 16.24 and 20.35 MJ. Considering the total mass of ejected coal, the average initial speed of ejected coal particles ranges from 24.98 to 27.96 m/s.

Estimated Value of Kinetic Energy of Rib Burst

Developing a protective structure on CM

Energy Analysis – A Protective Structure for CM

Numerical modellingNumerical Modelling of Dynamic Load

Numerical model of SHPB test system

Numerical model of Pre-load SHPB test system

Drop hammer test system

Influence of beddings on dynamic behaviour of coal- Numerical Simulation of SHPB Test with particle flow code (PFC)

19

Numerical model of SHPB test system

Numerical models of specimen (red represents beddings in coal specimen)

(a) S1 (b) S2 (c) S3 (d) S4

Split Hopkinson Pressure Bar (SHPB)

Stress wave propagation in bars with specimen S1 (no beddings): the red denotes tensile wave and the black denotes compressive wave

Stress Wave Propagation (resulting from dynamic load)

Failure Mode of Specimen

(a) S1 (b) S2 (c) S3 (d) S4

Failure evolution of different specimensFragment (fracture) pattern and failure

mode of each specimen at 1000us

• Specimen S1 and S2 have the most severe damage and similar failure modes, both characterised by shear failure.

• Specimen S3 basically remains intact with a little tensile cracks perpendicular to beddings.

Strain energy changes vs time for different specimens

Strain Energy

§ Beddings in a coal specimen lead to the degradation of its dynamic mechanical properties. This influence is closely associated with the angle between bedding and loads direction. When dynamic loads are inclined to beddings, specimen is most vulnerable with bedding breaking and sliding.

§ Strain energy and failure are effected by beddings. For specimen containing inclining beddings, coal bump and burst are not likely to appear in such coal as its instability is gradual and its storage capacity of strain energy is limited. Coal specimens with beddings parallel to dynamic loads is more vulnerable to burst.

Modelling of Water (moisture) influence Numerical model

Nuclear magnetic resonance (NMR)-images of sandstone disk with different water contents: a

saturation process; b drying process (Zhou, 2016)

NMR-images of sandstone disk in saturation condition

The relationship between saturation degree and distance ratio: (a) saturation

distribution; (b) evaporation distribution

The water distribution curve and numerical model (sc=0.3); the blue patterns represent water-weakened contacts and the green patterns represent normal contacts.

Comparison between experimental results of dry specimen and saturated specimen under uniaxial compression

Modelling of Water (moisture) influence Numerical simulation

Sketch of the numerical experimentFlow chart for the simulation procedure

Stress evolution versus ks in different cases

Instability water saturation coefficient for specimens in high-stress conditions

vs evolution curves with ks increasing

Initial stress coefficient:

65%~80% UCS: Lower instability point and higher coal burst risk.

40%~65% UCS: Water infusion is an effective approach to reduce rock burst risk as having been reported by many literatures.

≤40% UCS: Water has limited effect on releasing stress and energy for coal at such a low stress level.

Modelling of Water (moisture) influence Numerical simulation

Final failure patterns of all damaged specimens

Failure evolution of specimen in Case 5, ks=0.8

• Failure patterns were dominated by shear failure through the specimens.

• Higher initial axial stress indicates more severely with more cracks and fragments.

• Failure intensity highly depends on the release of strain energy.

Modelling of Water (moisture) influence Numerical simulation

• Fundamental studies of mechanisms

• A new risk assessment methodology for coal/rock burst;

• Monitoring and mitigation technologies.

Numerical modelling: Coal and Rock Burst

Thank you! Thank you!

Questions?

Contact

Ting Ren

tren@uow.edu.au

Questions？