BỘ GIÁO DỤC VÀ ĐÀO TẠO

TRƯỜNG ĐẠI HỌC SƯ PHẠM KỸ THUẬT

THÀNH PHỐ HỒ CHÍ MINH
KHOA: CƠ KHÍ CHẾ TẠO MÁY

Báo cáo quá trình môn Ứng dụng CAE trong thiết kế
- Nhóm 04CLC

Họ và tên: Đinh Ngọc Quý

MSSV: 22144175
GVHD: Trần Thái Sơn

TP. HCM, tháng 10 năm 2024

 Convergence Study of 3D Solid Elements

9.3.1 About the Cantilever Beam

9.3.3 Create a 3D Model in DesignModeler

9.3.4 Set Up Support, Load, and Solution Objects

9.3.5 Lower-Order Hexahedra
5.0
4.0
3.0
2.0
1.5
1.3
1.2
1.0
0.9
0.8
Element Size (mm) Number of Nodes Tip Deflection (mm)

5 625 0.2211

4 1116 0.22148

3 2624 0.22184

2 7381 0.22214

1.5 15876 0.22228

1.3 23808 0.22234

1.2 29189 0.22236

1 48000 0.22241

0.9 70886 0.22244

0.8 93750 0.22246

 Convergence curve
for lower-order
hexahedra

0.223
Tip Deflection (mm)

0.2225

0.222

0.2215

0.221

0.2205

0.22
0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0
1 00 2 00 3 00 4 00 5 00 6 00 7 00 8 00 9 00 00
10

Number of Nodes

9.3.6 Lower-OrderTetrahedra
5.0
4.0
2.5
2.0
1.5
1.2
1.0
0.8
0.7
0.6
0.55
0.5
Element Size (mm) Number of Nodes Tip Deflection (mm)

5 190 0.15389

4 304 0.16425

2.5 675 0.18127

2 1341 0.19696

1.5 1941 0.20091

1.2 2769 0.20307

1 5005 0.20541

0.8 6789 0.20907

0.7 8595 0.20949

0.6 12632 0.21061

0.55 14697 0.21053

0.5 19223 0.21117

Convergence curve for

lower-order tetrahedra
 0.25
Tip Deflection (mm)

0.2

0.15

0.1

0.05

0
0 5000 10000 15000 20000 25000

Number of Nodes

9.3.7 Lower-Order Prisms (Parallel to Loading Direction)

5.0
4.0
3.0
2.4
2.0
1.5
1.2
1.0
Element Size (mm) Number of Nodes Tip Deflection (mm)
5 625 0.21362

4 1116 0.21636

3 2920 0.21888

2.4 4536 0.21997

2 8008 0.22072

1.5 16996 0.22144

1.2 32436 0.22179

1 54740 0.22199

Convergence curve for

lower-order parallel
prisms
 Tip Deflection (mm) 0.224

0.222

0.22

0.218

0.216

0.214

0.212

0.21

0.208
0 10000 20000 30000 40000 50000 60000

Number of Nodes

9.3.8 Lower-Order Prisms (Perpendicular to Loading

Direction)
5.0
4.0
3.0
2.4
2.0
1.5
1.2
1.0
Element Size (mm) Number of Nodes Tip Deflection (mm)

5 625 0.19175

4 1116 0.20192

3 2920 0.21311

2.4 4545 0.21507

2 8008 0.21749

1.5 16996 0.21937

1.2 32436 0.22055

1 54740 0.22115
 Convergence curve for
lower-order perpendicular
prisms

0.225

0.22
Tip Deflection (mm)

0.215

0.21

0.205

0.2

0.195

0.19

0.185

0.18

0.175
0 10000 20000 30000 40000 50000 60000

Number of Nodes

9.3.9 Higher-Order Hexahedra

5.0
4.0
3.0
2.0

1.5
1.3
Element Size (mm) Number of Nodes Tip Deflection (mm)

5 2225 0.22203

4 4056 0.22217

3 9776 0.22231

2 28061 0.22242

1.5 61040 0.22247

1.3 92000 0.22249

 Convergence
curve for higher-
order hexahedra
0.2226
Tip Deflection (mm)

0.2224

0.2222

0.222

0.2218

0.2216

0.2214

0.2212

0.221
0 2000 4000 6000 8000 10000 12000 14000

Number of Nodes

9.3.10 Higher-OrderTetrahedra
5.0
4.0
3.0
2.5
2.0
1.5
1.2
1.0
0.8
0.7
Element Size (mm) Number of Nodes Tip Deflection (mm)
5 968 0.22154

4 1699 0.22183

3 3123 0.22206

2.5 3854 0.22211

2 8021 0.22225

1.5 11684 0.22236

1.2 16772 0.2224

1 31137 0.22245

0.8 42298 0.22249

0.7 53749 0.2225

Convergence curve
for higher-order
tetrahedra
 0.2226
Tip Deflection (mm)

0.2224

0.2222

0.222

0.2218

0.2216

0.2214

0.2212

0.221
0 10000 20000 30000 40000 50000 60000

Number of Nodes

9.3.11 Higher-Order Parallel Prisms

5.0
4.0
3.3

3.0
2.4
2.0
1.8
1.6
1.4
Element Size (mm) Number of Nodes Tip Deflection (mm)

5 2705 0.22194

4 4956 0.22211

3.3 8523 0.22222

3 13459 0.22227

2.4 21105 0.22234

2 37739 0.2224

1.8 50012 0.22242

1.6 69699 0.22244

1.4 104137 0.22247

 Convergence curve
for higher-order
parallel prisms

0.2226
Tip Deflection (mm)

0.2224

0.2222

0.222

0.2218

0.2216

0.2214

0.2212

0.221
0 5000 10000 15000 20000 25000 30000 35000 40000 45000

Number of Nodes

9.3.12 Higher-Order Perpendicular Prisms

5.0

4.0
3.3
3.0
2.4
2.0
1.8
1.6
1.4
Element Size (mm) Number of Nodes Tip Deflection (mm)

5 2705 0.22183

4 4956 0.22203

3.3 8523 0.22216

3 13459 0.22222

2.4 21149 0.2223

2 37739 0.22237

1.8 50012 0.2224

1.6 69699 0.222442

1.4 104137 0.22245

 Convergence curve
for higher-order
perpendicular prisms

0.2226
Tip Deflection (mm)

0.2224

0.2222

0.222

0.2218

0.2216

0.2214

0.2212

0.221
0 5000 10000 15000 20000 25000 30000 35000 40000 45000

Number of Nodes

9.3.13 Comparison: Lower-Order Elements

The chart below is made by a collection of the convergence curves in 9.3.5 to 9.3.8
to compare the convergence behaviors of the lower-order elements. The order of
the convergence speed is, from fast to slow, hexahedron, parallel prism,
perpendicular prism, and tetrahedron. The differences among them are obvious.
The lower- order tetrahedron converges so poorly that it is not practically useful.
As a guideline, NEVER use lower-order tetrahedral elements.

Lower-order
parallel prisms

0.25
Tip Deflection (mm)

0.2

Lower-order
0.15 perpendicular
Lower-
prisms order
hexa-
0.1 Lower- hedra
order
tetrahedra
0.05

0
0 10000 20000 30000 40000 50000 60000

Number of Nodes

9.3.14 Comparison: Higher-Order Elements

The chart below is made from a collection of the convergence curves in 9.3.9 to
9.3.12 to compare the convergence behaviors of the higher-order elements. The
differences among them are not obvious but still distinguishable. In contrast to the
lower-order tetrahedron, the higher-order tetrahedron is still practically useful as
long as the mesh is fine enough.

Higher-order
Higher-order hexahedra
tetrahedra
0.2226

0.2224
Tip Deflection (mm)

0.2222

0.222
Higher-order
Higher- perpendicula
0.2218
order r prisms
parallel
0.2216
prisms
0.2214

0.2212

0.221
0 20000 Number
40000 of Nodes
60000 80000 100000 120000

9.3.15 Comparison: Hexahedra

The chart below is made by a collection of the convergence curves in 9.3.5 and
9.3.9 to compare the convergence behaviors between the lower- and higher-order
hexahedra. It is obvious that the higherorder hexahedral element is better than the
lower-order hexahedral, but the difference is not so dramatic as the tetrahedral
element (see 9.3.16, this page).

Higher-order
hexahedra

Lower-order
hexahedra

9.3.16 Comparison:Tetrahedra
The chart below is made from a collection of the convergence curves in 9.3.6 and
9.3.10 to compare the convergence behaviors between the lower- and higher-order
tetrahedra. It is obvious that higher-order element is much better than the lower-
order one. Remember: NEVER use lower-order tetrahedral elements.

0.25
Tip Deflection (mm)

0.2

0.15 Higher-order
Lower-order tetrahedra
tetrahedra
0.1

0.05

0
0 10000 20000 30000 40000 50000 60000

Number of Nodes

9.3.17 Comparison: Parallel Prisms

The chart below is made by a collection of the convergence curves in 9.3.7 and
9.3.11 to compare the convergence behaviors between the lower- and higher-order
parallel prisms. It is obvious that higherorder element is much better than the
lower-order one. Like lower-order tetrahedral, lower-order prismatic elements are
not recommended.

0.2226

0.2225
Tip Deflection (mm)

0.2224

0.2223
Higher-order
0.2222 parallel
0.2221
Lower-order prisms
parallel prisms
0.222

0.2219

0.2218

0.2217

0.2216
0 20000 40000 60000 80000 100000 120000

Number of Nodes

9.3.18 Comparison: Perpendicular Prisms

The chart below is made by a collection of the convergence curves in 9.3.8 and
9.3.12 to compare the convergence behaviors between the lower- and higher-order
perpendicular prisms. It is obvious that the higher-order element is much better
than the lower-order one. Like lower-order tetrahedral, lower-order prismatic
elements are not recommended.

0.2225

0.2224
Tip Deflection (mm)

0.2223

0.2222

0.2221 Higher-order
Lower-order perpendicular
0.222
perpendicular prisms
0.2219 prisms
0.2218

0.2217

0.2216

0.2215
0 20000 40000 60000 80000 100000 120000

Number of Nodes

9.3.19 Summary and Guidelines

Combining the observations in Section 3.5 and this section, we may
summarize the conclusions as follows: (a) Never use lower-order tetrahedra or
triangles, (b) Higher-order tetrahedra or triangles are as good as other elements as
long as the mesh is fine enough. In cases of coarse mesh, however, they perform
poorly and are not recommended, (c) Lower-order prisms are not recommended,
(d) Lower-order hexahedra and quadrilaterals can be used, but they are not as
efficient as their higher-order counterparts, (e) Higher-order hexahedra, parallel-
prisms, and quadrilaterals are among the most efficient elements we have discussed
so far. Mesh your models with these elements whenever possible. If that is not
possible, then at least try to achieve a higher-order hexahedra-dominant or
quadrilateraldominant mesh. |
Remark: CPU Time [2] In Section 3.5 and this section, comparisons among
elements are made under the same number of nodes. More reasonable comparisons
should be made under the same CPU time. For a simulation task, the CPU time
consists of three parts. First, the time required to establish Eq. 1.3.1 . It may
involve numerical integrations for each element. This part of CPU time depends on
the total number of elements as well as the number of integration points of each
element. Second, the time required to solve the equation. This part of CPU time is
determined solely by the number of degrees of freedom, which is in turn
determined by the number of nodes and the dimensionality (2D or 3D). Third, the
others (housekeeping, overhead, etc). For small problems, the overall CPU time is
dominated by the third part. That is why we didn't use CPU time for comparison,
since all cases are small when coarsely meshed. For large problems, the third part
is negligible and the CPU time is essentially the sum of the first two parts.
Therefore, strictly speaking, our comparison was not perfectly accurate.
Nevertheless, the discussions and conclusions in this section pretty much reflect
the reality. These guidelines should be useful. |