Finite Element Analysis of Stress Distribution in Underground Galleries with Varying Dimensions

Finite Element Analysis of Stress Distribution in Underground Galleries with Varying Dimensions

Mustafa Emre Yetkin* Muharrem Kemal Özfırat Hayati Yenice Turgay Onargan


Corresponding Author Email: 
mustafa.yetkin@deu.edu.tr
Page: 
20-27
|
DOI: 
https://doi.org/10.18280/eesrj.110102
Received: 
21 November 2023
|
Revised: 
11 January 2024
|
Accepted: 
18 January 2024
|
Available online: 
31 March 2024
| Citation

© 2024 The authors. This article is published by IIETA and is licensed under the CC BY 4.0 license (http://creativecommons.org/licenses/by/4.0/).

OPEN ACCESS

Abstract: 

In underground mining, stresses within the rock mass are initially in a state of equilibrium. Excavation disrupts this balance, necessitating the establishment of a new stress equilibrium. The reinforcement of mine galleries is contingent upon an understanding of the geomechanical properties of the rock mass. Furthermore, the dimensions of these galleries significantly influence the distribution of stresses around them. This study investigates the normal and horizontal stresses surrounding galleries of varying dimensions, excavated under identical rock mass conditions. The findings facilitate the recommendation of optimal gallery sizes that minimize stress concentrations. Consequently, these recommendations are instrumental in enabling the selection of gallery sizes that enhance mining safety and ensure sustainable production.

Keywords: 

finite element method, stress distribution, underground galleries, mining

1. Introduction

The design and execution of mine galleries are pivotal to mining operations, underpinning critical functions such as occupational safety, ventilation, water management, and the movement of personnel and materials. Appropriately dimensioned galleries are imperative for a productive and secure work environment. The determination of these dimensions is influenced by a myriad of factors, including underground stress conditions, the transportation demands for heavy machinery, and the operational requirements for personnel and equipment. Informed selection of gallery dimensions, guided by these parameters, is essential for optimizing occupational safety and operational efficiency.

Historically, research efforts have been directed toward modeling stresses and failure potentials associated with circular cross-section tunnels within rock masses. The response of these underground structures to static loads has been extensively examined through numerical simulations [1-6].

In the present study, gallery models measuring 4×4 meters and 5×5 meters were developed within an identical geological setting. The selection process for the gallery dimensions was informed by the strength characteristics of the local rock mass and the prevalent dimensions within existing mining operations. Consequently, dimensions of 4×4 and 5×5 meters were identified as optimal for further investigation. The stress profiles, both vertical and horizontal, associated with these gallery dimensions were scrutinized. Initially, a representative model of the field was constructed, enabling the estimation of stress distributions. Subsequently, the computed stress values were imposed upon the gallery models to determine the stress patterns that would manifest around these structures. Based on the stress distributions obtained, recommendations pertaining to the most advantageous gallery dimensions were formulated.

2. Creation of the Model

The data of the rock material considered in the created field model are provided in Table 1. The underground quarry under investigation in this study is located 70 km east of İzmir and 15 km northwest of Bayındır.

Table 1. Data related to rock material [7]

Rock Material

Deadrock

Uniaxial Compressive Strength

UCS (MPa)

35.67

Geological Strength Index

GSI

60

Rock Mass Constant

mi

20

Density

ɣ (t/m3)

2.7

During the creation of the field model, a depth of 90 meters from the surface was considered for excavating the galleries. In the subsequent stage, using the rock material properties provided in Table 1 and considering the depth, the parameters related to the rock mass were calculated using RocData software [8]. The process of obtaining the rock mass data is illustrated in Figure 1.

Figure 1. Obtaining rock mass data

The data related to the rock mass calculated using the RocData software are provided in Table 2.

Table 2. Rock mass data used in the model

Rock Mass

Deadrock

Elastic Modulus

EM (MPa)

6955.650

Cohesion

C (MPa)

0.752

Tensile Strength

σt (MPa)

0.087

Internal Friction Angle

ο (Derece)

54.950

Density

ɣ (t/m3)

2.700

Poisson's Ratio

ν

0.300

The horizontal and vertical stress values that will occur at a working depth of 90 meters under the selected rock mass conditions, considering the Poisson's ratio of the dominant rock structure in the field, have been calculated using the following equations [9]:

$\sigma \mathrm{h}=\mathrm{k} . \sigma \mathrm{v}$            (1)

$\mathrm{k}=0,25+7 . \mathrm{E} .(0,001+1 / \mathrm{z})$            (2)

Here is a brief explanation of the variables: σv: Vertical stress in megapascals (MPa); σh: Horizontal stress in megapascals (MPa); k: Ratio of horizontal stress to vertical stress (no units); E: Average modulus of elasticity of the rock mass in the vertical direction up to the working depth in gigapascals (GPa); z: Working depth in meters (m).

The vertical stress at a working depth of 90 m was calculated as 2.33 MPa using Eq. (1). Considering the rock mass properties given in Table 2 and the calculated coefficient k as 0.84 based on the rock mass ’s modulus of elasticity and depth (Eq. (2)), the expected horizontal stress in the field was calculated as 1.96 MPa. The calculated coefficient k was incorporated into the field model, ensuring an accurate representation of stress conditions within the rock mass environment where the galleries will be excavated. Field models with dimensions of 4×4 meters and 5×5 meters were created using Phase2D software [10]. A visual representation of the models is provided in Figure 2.

Figure 2. Different gallery size models

After entering the data the models, the models were run, and normal and horizontal stresses around the galleries were calculated for different gallery sizes. The stresses around the galleries with gallery sizes of 4×4 and 5×5 meters in the same rock mass environment are shown in Figure 3.

Figure 3. Normal stresses occurring around galleries of different sizes

The calculated stress values are detailed in Table 3. Normal stresses occurring immediately above the gallery were measured along an approximately 14-meter line. The horizontal stress values on the right and left walls were recorded by measurements along an approximately 4-meter line. The measurements were conducted using measurement lines placed on models established within the software. These measurement lines are structures permitted to be generated by the software, allowing the user to obtain results of their desired intervals and types in the regions they pass through. The measurements were taken just above the gallery and on the side walls.

Table 3. Normal and horizontal stresses occurring around galleries of different sizes

Gallery Size

4×4 meter

5×5 meter

4×4 meter

5×5 meter

Top of Gallery

Horizontal Stresses (MPa)

Normal Stresses (MPa)

Left Wall

Right Wall

Left Wall

Right Wall

3.321

3.194

-0.325

-0.360

-0.183

-0.221

3.764

4.103

-0.201

-0.227

-0.102

-0.130

4.140

4.425

-0.077

-0.095

-0.021

-0.038

4.425

4.607

0.047

0.037

0.061

0.053

4.570

4.687

0.172

0.170

0.142

0.145

4.771

4.842

0.397

0.399

0.362

0.368

4.814

4.927

0.724

0.726

0.720

0.723

4.910

4.917

1.052

1.052

1.078

1.078

4.902

4.919

1.379

1.378

1.436

1.433

4.878

4.893

1.706

1.705

1.794

1.788

4.807

4.819

Mean

4.755

4.770

0.487

0.479

0.529

0.520

4.679

4.698

       

4.629

4.638

       

4.578

4.580

       

4.510

4.521

       

4.452

4.469

       

4.417

4.429

       

4.384

4.396

       

4.370

4.378

       

4.361

4.371

       

4.370

4.376

       

4.385

4.397

       

4.418

4.428

       

4.456

4.468

       

4.510

4.519

       

4.574

4.578

       

4.626

4.634

       

4.677

4.685

       

4.754

4.765

       

4.804

4.815

       

4.867

4.872

       

4.881

4.911

       

4.906

4.898

       

4.807

4.929

       

4.778

4.823

       

4.581

4.689

       

4.416

4.649

       

4.124

4.478

       

3.732

4.133

       

3.325

3.134

       

Mean

       

4.498

4.555

       
3. Creating Gallery Models

In the context of modeling studies, two galleries have been designed with dimensions of 4×4 meters and 5×5 meters, both with a length of 20 meters. The gallery designs were created using the SpaceClaim [11] solid modeling software. A visual representation of the created models is provided in Figure 4. The purpose of creating three-dimensional models is to examine the normal stresses occurring in the third dimension along the z-axis around the galleries in more detail, as the models created in Phase2D software are two-dimensional.

Figure 4. The created gallery models

After the creation stage of the gallery models, the models were transferred to the ANSYS [12] stress analysis program. The values provided in Table 2 were used as input parameters for the models in the ANSYS stress analysis program, thus ensuring that the model behaves like a real rock environment. The process of entering rock mass (Elastic Modulus, Density and Poisson's Ratio) data into the models is shown in Figure 5.

Figure 5. Entering input parameters into the gallery models

An image of the gallery models created in the ANSYS stress analysis program is provided in Figure 6.

Figure 6. Gallery models in the 3rd dimension

The average normal stress values applied to the gallery models obtained from the Phase2D program and occurring on the gallery are as follows:

For the 4×4 meter gallery:

Average normal stress on the gallery: 4.498 MPa, Average horizontal stress on the right wall of the gallery: 0.479 MPa, Average horizontal stress on the left wall of the gallery: 0.487 Mpa.

For the 5×5 meter gallery:

Average normal stress on the gallery: 4.555 MPa, Average horizontal stress on the right wall of the gallery: 0.520 MPa, Average horizontal stress on the left wall of the gallery: 0.529 MPa.

These calculated stress values were applied regionally to the gallery models using the ANSYS stress analysis program. Before proceeding with stress analysis, the models were fixed to the ground and the effect of gravity was applied to the models. This was done to ensure that the models operate as accurately as possible. A visual representation of the models after applying the loads is provided in Figure 7.

Figure 7. Application of normal (vertical) and horizontal (horizontal) stresses to gallery models

After completing these steps, the stress analysis process for the models was executed. An image showing the stress distribution around galleries of different sizes is provided in Figure 8. For the 4×4 meter gallery, the maximum vertical deformation was calculated to be 15.71 mm, whereas for the 5×5 meter gallery, it was calculated to be 19.85 mm. The distribution of deformations around the gallery is shown in Figure 9 and Figure 10. It is observed that the maximum deformation occurs in the central regions located just above the gallery in both cases.

Figure 8. Stress distribution

Figure 9. Vertical deformations occurring in the 4×4 meter gallery size

Figure 10. Vertical deformations occurring in the 5×5 meter gallery size

In the ANSYS stress analysis, the combined and graphical representations of the normal stresses occurring around galleries of different sizes are provided in Table 4 and Figure 11.

Table 4. Normal stresses occurring around galleries of different sizes

Gallery Size

4×4 meter

5×5 meter

Left Wall

Right Wall

Top of Gallery

Left Wall

Right Wall

Top of Gallery

Normal Stresses (MPa)

24.973

24.991

16.322

27.986

28.089

17.075

24.869

24.890

16.444

28.178

28.276

17.543

24.861

24.885

16.567

28.450

28.545

18.013

24.571

24.597

16.438

28.315

28.408

17.949

23.571

23.596

16.310

27.256

27.349

17.887

22.597

22.622

16.047

26.229

26.323

17.711

21.951

21.973

15.785

25.462

25.557

17.535

21.467

21.485

15.449

24.821

24.918

17.298

20.991

21.005

15.113

24.192

24.289

17.061

20.780

20.789

14.778

23.838

23.936

16.789

20.573

20.578

14.445

23.491

23.589

16.517

20.403

20.404

14.165

23.194

23.292

16.255

20.299

20.296

13.887

22.990

23.089

15.994

20.199

20.193

13.687

22.792

22.890

15.783

20.131

20.121

13.488

22.643

22.742

15.572

20.080

20.068

13.371

22.522

22.621

15.437

20.033

20.018

13.254

22.405

22.504

15.302

20.003

19.987

13.216

22.331

22.430

15.255

19.977

19.960

13.178

22.261

22.359

15.209

19.956

19.937

13.216

22.202

22.300

15.255

19.941

19.921

13.254

22.162

22.260

15.302

19.929

19.909

13.371

22.126

22.224

15.437

19.921

19.900

13.488

22.104

22.203

15.572

19.916

19.895

13.687

22.092

22.190

15.783

19.915

19.894

13.887

22.083

22.181

15.994

19.916

19.895

14.165

22.092

22.190

16.255

19.921

19.900

14.445

22.104

22.203

16.517

19.929

19.909

14.778

22.126

22.224

16.789

19.941

19.921

15.113

22.162

22.260

17.061

19.956

19.937

15.449

22.202

22.300

17.298

19.977

19.960

15.785

22.261

22.359

17.535

20.003

19.987

16.047

22.331

22.430

17.711

20.033

20.018

16.310

22.405

22.504

17.887

20.080

20.068

16.438

22.522

22.621

17.949

20.131

20.121

16.567

22.643

22.742

18.013

20.199

20.193

16.444

22.792

22.890

17.543

20.299

20.296

16.322

22.990

23.089

17.075

20.403

20.404

 

23.194

23.292

 

20.573

20.578

 

23.491

23.589

 

20.780

20.789

 

23.838

23.936

 

20.991

21.005

 

24.192

24.289

 

21.467

21.485

 

24.821

24.918

 

21.951

21.973

 

25.462

25.557

 

22.597

22.622

 

26.229

26.323

 

23.571

23.596

 

27.256

27.349

 

24.571

24.597

 

28.315

28.408

 

24.861

24.885

 

28.450

28.545

 

24.869

24.890

 

28.178

28.276

 

24.973

24.991

 

27.986

28.089

 

Mean

21.304

21.305

14.884

24.126

24.223

16.680

Figure 11. The graphical representation of the stresses

4. Results and Discussion

In the scope of this study, vertical stresses expected to occur around galleries of different sizes were calculated through two different stress analyzes in the first stage, the vertical and horizontal stresses that would affect the galleries were calculated using the Phase2D program. In the second stage, the ANSYS stress analysis software was used to measure the vertical stress values around the gallery and the deformation and vertical stress values that would occur in the third dimension within the galleries. Because of the stress analyzes conducted in the first stage, it can be observed from the values given in Table 3 that the stresses increase linearly from the base region of the gallery toward the upper region of the gallery. They decrease slightly, stabilizing, and are influenced by the pressure arch formed on the gallery's upper portion. This situation has also been explained in detailed scientific studies on the subject [13-20]. It is noted that the vertical stresses occurring in the 4×4 meter and 5×5 meter gallery sizes are very close to each other, but the vertical stresses on the gallery's upper portion are particularly lower in the 4×4 meter gallery size.

Looking at the average horizontal stress values around the gallery given in Table 3 because of the first-stage stress analysis, it is observed that the stresses on the left and right walls of the 4×4 meter and 5×5 meter gallery sizes are very close to each other. The lowest average vertical stress value was observed in the upper region of the 4×4 meter gallery size.

Because of the two-stage stress analyzes it was determined that the horizontal stress values occurring in the left and right regions of the gallery are very close to each other. However, especially in the graphical representation shown in Figure 11, it is evident that the normal stresses occurring on the gallery and analyzed in the second and third dimensions are lower in the 4×4 meter gallery size. In addition, it allows for less deformation. Therefore, it has been determined that designing the gallery sizes to be opened in the working area as 4×4 meters would be suitable in terms of both work efficiency and health.

5. Conclusions

Predicting the behavior of the rock mass in which galleries will be excavated under load is crucial for gallery stability, work efficiency, and safety. Galleries that are excavated in the correct dimensions will result in a more balanced stress distribution, ensuring the stability of the gallery throughout the mining operation. In the scope of this study, galleries designed in sizes of 4×4 meters and 5×5 meters, which are planned to be excavated within the considered rock mass, were analyzed for the stresses occurring around them through three-dimensional analysis.

It is undeniable that three-dimensional models better reflect the mining conditions. According to the obtained results, it was observed that the stresses and vertical deformation values are lower in the 4×4 meter gallery size. Considering the geomechanical parameters of the rock mass and the results of stress analysis, no failure is expected in galleries to be excavated in these dimensions. However, through regular discontinuity control in the field, precautions can be taken to prevent ceiling collapses due to the formation of rock wedges during gallery excavation.

  References

[1] He, B., Yin, G.Z. (2014). Stability analysis of tunnel during the excavation based on ANSYS. Applied Mechanics and Materials, 577: 1135-1138. https://doi.org/10.4028/www.scientific.net/AMM.577.1135

[2] Fakhimi, A., Carvalho, F., Ishida, T., Labuz, J.F. (2002). Simulation of failure around a circular opening in rock. International Journal of Rock Mechanics and Mining Sciences, 39(4): 507-515. https://doi.org/10.1016/S1365-1609(02)00041-2

[3] Hendron, A.J., Engeling, P.D., Paul, S.L., Aiyer, A.K. (1972). Geomechanical model study of the behavior of underground openings in rock subjected to static loads: Report 3: Tests on lined openings in jointed and intact rock. Technical Report N-69-1. Waterways Experiment Station, Vicksburg, Mississippi, US.

[4] Huang, F., Zhu, H., Xu, Q., Cai, Y., Zhuang, X. (2013). The effect of weak interlayer on the failure pattern of rock mass around tunnel-scaled model tests and numerical analysis. Tunnelling and Underground Space Technology, 35: 207-218. https://doi.org/10.1016/j.tust.2012.06.014

[5] Ren, F., Chang, Y., He, M. (2020). A systematic analysis method for rock failure mechanism under stress unloading conditions: A case of rock burst. Environmental Earth Sciences, 79: 1-16. https://doi.org/10.1007/s12665-020-09111-2

[6] Yetkin, M.E. (2022). Metalik Madenlerde Topuk Boyutlarındaki Değişimin Gerilme Dağılımına Etkisi. Afyon Kocatepe Üniversitesi Fen Ve Mühendislik Bilimleri Dergisi, 22(6): 1459-1468. https://doi.org/10.35414/akufemubid.1148884

[7] Mallı, T., Yetkin, M.E., Özfırat, M.K., Kahraman, B.A. Y.R.A.M. (2017). Numerical analysis of underground space and pillar design in metalliferous mine. Journal of African Earth Sciences, 134: 365-372. https://doi.org/10.1016/j.jafrearsci.2017.07.018

[8] RocData. (2014). Rock, Soil and Discontinuity Strength Analysis, Version 5.0. https://www.rocscience.com/software/rsdata#.

[9] Sheorey, P.R. (1994). A theory for in Situ stresses in isotropic and transverseley isotropic rock. International Journal of Rock Mechanics and Mining Sciences & Geomechanics Abstracts, 31(1): 23-34. https://doi.org/10.1016/0148-9062(94)92312-4

[10] Phase2, Version 8.020. (2014). Rocscience Inc, Toronto, Ontario, Canada. https://www.rocscience.com/software/rs2.

[11] SpaceClaim Corporation. (2014). SpaceClaim (2014.0.1.03201), SpaceClaim Corporation. https://www.ansys.com/products/3d-design/ansys-spaceclaim.

[12] Ansys Mechanical. (2018). ANSYS, Inc., Canonsburg PA, USA. https://www.ansys.com/products/structures/ansys-mechanical.

[13] Poulsen, B.A. (2010). Coal pillar load calculation by pressure arch theory and near field extraction ratio. International Journal of Rock Mechanics and Mining Sciences, 47(7): 1158-1165. https://doi.org/10.1016/j.ijrmms.2010.06.011

[14] Wang, S.L., Hao, S.P., Chen, Y., Bai, J.B., Wang, X. Y., Xu, Y. (2016). Numerical investigation of coal pillar failure under simultaneous static and dynamic loading. International Journal of Rock Mechanics and Mining Sciences, 84: 59-68. https://doi.org/10.1016/j.ijrmms.2016.01.017

[15] Basarir, H., Oge, I.F., Aydin, O. (2015). Prediction of the stresses around main and tail gates during top coal caving by 3D numerical analysis. International Journal of Rock Mechanics and Mining Sciences. 76: 88-97. https://doi.org/10.1016/j.ijrmms.2015.03.001

[16] Deng, G., Xie, H., Gao, M. (2023). Fracture mechanisms of competent overburden under high stress conditions: A case study. Rock Mechanics and Rock Engineering, 56: 1759-1777. https://doi.org/10.1007/s00603-022-03169-z

[17] Yetkin, M.E., Özfırat, M.K., Onargan, T. (2024). Examining the optimum panel pillar dimension in longwall mining considering stress distribution. Scientific Reports, 14(6928): 1-12. https://doi.org/10.1038/s41598-024-57579-w

[18] Kang, H., Wu, L., Gao, F. (2019). Field study on the load transfer mechanics associated with longwall coal retreat mining. International Journal of Rock Mechanics and Mining Sciences, 124: 104141. https://doi.org/10.1016/j.ijrmms.2019.104141

[19] Sinha, S., Walton, G. (2019). Investigation of longwall headgate stress distribution with an emphasis on pillar behavior. International Journal of Rock Mechanics and Mining Sciences, 121: 104049. https://doi.org/10.1016/j.ijrmms.2019.06.008

[20] Li, C.C. (2006). Rock support design based on the concept of pressure arch. International Journal of Rock Mechanics and Mining Sciences, 43(7): 1083-1090. https://doi.org/10.1016/j.ijrmms.2006.02.007