You do not have permission to edit this page, for the following reason:

You are not allowed to execute the action you have requested.


You can view and copy the source of this page.

x
 
1
<!-- metadata commented in wiki content
2
3
4
==An improved FEM-DEM coupling simulation for granular-media-based thin-wall elbow tube push-bending process==
5
6
Hai Liu<span id="fnc-1"></span><span style="text-align: center; font-size: 75%;">[[#fn-1|<sup>1</sup>]]</span>, Guang-Hui Ma, Zhe Geng
7
8
Department of Precision Manufacturing Engineering, Suzhou Vocational Institute of Industrial Technology, 1 Zhineng Avenue, Suzhou 215104, P.R. China.
9
-->
10
==Abstract==
11
12
The granular-media-based push-bending process has been developed to manufacture thin-wall elbow tube with t/D<math>\le 0.01</math> (the ratio of wall thickness to outer diameter) and R/D<math>\le 1.5</math> (the ratio of bending radius to outer diameter). In the process, a tubular blank is filled with granular media and then pushed into a die to form an elbow shape. To investigate the process, a FEM-DEM coupling model has been developed, in which FEM is used to simulate bending deformation of tubular blank, and DEM is used to calculate contact forces between spherical particles in granular media. In this work, an improved numerical formulation is proposed in order to reach mechanical equilibrium quickly and accelerate the convergence of DEM simulation, when the new contacts are no longer created and the old contacts are no longer deleted in granular media. Using the proposed numerical formulation, the improved FEM-DEM coupling simulation for granular-media-based thin-wall elbow tube push-bending process is less time-consuming than before under the same simulation condition.
13
14
'''Keywords:''' Elbow tube, granular-media-based, push bending, energetic formulation, FEM-DEM coupling
15
16
==1. Introduction==
17
18
The thin-wall elbow tubes with t/D≤0.01 and R/D<math>\le 1.5</math> (where t is wall thickness, D is outer diameter, and R is bending radius) are widely used in rocket engine pipelines [1-3]. In the bending process of elbow tube with quite small t/D and R/D, the forming defects such as wrinkling and fracture is difficult to be avoided, and the dimensional precision is difficult to satisfy the requirements due to the springback and cross-sectional ovalization. Push-bending process is demonstrated to be applicable for manufacturing elbow tubes with small R/D [4-5]. The recently developed granular-media-based push-bending process can be used to manufacture the thin-wall elbow tube with t/D<math>\le 0.01</math> and R/D<math>\le 1.5</math> as required [6]. The granular-media-based thin-wall elbow push-bending process involves filling a tube with spherical granular media and pushing the tube into a die to bend a tubular blank into an elbow shape. The spherical granular media introduced as filler has great potential for its flexibility [7], resistance to high forming temperature [8-9], and characteristics of pressure-transmission [10-11]. In addition, the spherical granular media is controllable and predictable because of its regular shape.
19
20
To investigate the granular-media-based thin-wall elbow push-bending process, a 3D FEM-DEM coupling model is developed [12]. In the coupling model, FEM is used to simulate bending deformation of tubular blank (continuum), and DEM is used to calculate contact forces between spherical particles in granular media (discrete media). DEM is a dynamic method, and damping is applied to the equations of motion in order to absorb vibration energy and reach mechanical equilibrium. The biggest time overhead in the coupling model is absorbing vibration energy in DEM simulation to reach mechanical equilibrium. To accelerate the convergence of DEM simulation, a static energetic method is proposed in the present paper.
21
22
Generally in the earlier stage of DEM simulation, some new contacts between pairs of particles are detected and created, and some old contacts are broken and deleted. But in the later stage of DEM simulation, the new contacts are no longer detected and created, the old contacts are no longer broken and deleted, and a static energetic formulation can be used to determine mechanical equilibrium.
23
24
==2. Mechanical model==
25
26
The granular-media-based push-bending process is illustrated in [[#img-1|Figure 1]]. A tubular blank with predetermined dimensions and filled with granular media is inserted into the die and then deformed according to the shape of the die. The tubular blank is bended under the forming force of punch, constraint force of die, and interaction of granular filler. The forming force of punch and constraint force of die is simulated with FEM. The contact forces between neighbor particles in granular media is simulated with DEM.
27
28
<div id='img-1'></div>
29
{| class="wikitable" style="margin: 0em auto 0.1em auto;border-collapse: collapse;width:60%;" 
30
|-style="background:white;"
31
|style="text-align: center;padding:10px;"| [[Image:Review_294821678474-image1-c.png|378px]]
32
|-
33
| style="background:#efefef;text-align:left;padding:10px;font-size: 85%;"| '''Figure 1'''. Illustration of the granular-media-based thin-wall elbow push-bending process. (a) Before bending. (b) Bending
34
|}
35
36
===2.1 Kinematics===
37
38
In the DEM simulation, when the new contacts are no longer created and the old contacts are no longer deleted in granular media, the positions of the spherical particles are collected in the <math> 3\times 1 </math> column matrixes:
39
40
{| class="formulaSCP" style="width: 100%; text-align: center;" 
41
|-
42
| 
43
{| style="text-align: center; margin:auto;" 
44
|-
45
| <math>{\bf x}_{i}={\left[ \begin{matrix}{x}_{i1}&{x}_{i2}&{x}_{i3}\end{matrix}\right] }^{T}</math>
46
|}
47
| style="width: 5px;text-align: right;white-space: nowrap;" | (1)
48
|}
49
50
where <math display="inline">x_{il}</math> denotes the <math display="inline">l</math>-th position component of the <math display="inline">i</math>-th particle. The spherical particles in the granular media are labeled by the indexes <math display="inline">i=1, 2, 3\ldots N</math>. The symbol <math>[ \,\, ]^T</math> denotes the transpose of a matrix.
51
52
In the same way, the centroid displacements and the rotation components of particles are collected in the <math> 3\times 1 </math> column matrixes:
53
54
{| class="formulaSCP" style="width: 100%; text-align: center;" 
55
|-
56
| 
57
{| style="text-align: center; margin:auto;" 
58
|-
59
| <math>\begin{matrix}{\bf v}_{i}={\left[ \begin{matrix}{v}_{i1}&{v}_{i2}&{v}_{i3}\end{matrix}\right] }^{T}\\\, {\bf w}_{i}={\left[ \begin{matrix}{w}_{i1}&{w}_{i2}&{w}_{i3}\end{matrix}\right] }^{T}\end{matrix}\,</math>
60
|}
61
| style="width: 5px;text-align: right;white-space: nowrap;" | (2)
62
|}
63
64
where <math>v_{il}</math> denotes the <math>l</math>-th displacement component of the <math>i</math>-th particle, and <math>w_{il}</math> denotes the <math>l</math>-th rotation component of the <math>i</math>-th particle.
65
66
The contact <math>k_c</math> formed by the particle <math>i</math> and the particle <math>j</math>, as shown in [[#img-2|Figure 2]].
67
68
<div id='img-2'></div>
69
{| class="wikitable" style="margin: 0em auto 0.1em auto;border-collapse: collapse;width:60%;" 
70
|-style="background:white;"
71
|style="text-align: center;padding:10px;"| [[Image:Review_294821678474-image2.png|500px]]
72
|-
73
| style="background:#efefef;text-align:left;padding:10px;font-size: 85%;"| '''Figure 2'''. Relation between normal overlap and displacements of a pair of spherical particles in contact. The dashed circles denote the initial positions and the solid circles denote the positions after loading
74
|}
75
76
77
Assuming that elastic deformations are localized in very small neighborhoods of the contact points, so that, the relative displacement between the rigid part of the two particles near the contact point <math display="inline">{\bf x}^{{(k}_{c})}</math> is:
78
79
{| class="formulaSCP" style="width: 100%; text-align: center;" 
80
|-
81
| 
82
{| style="text-align: center; margin:auto;" 
83
|-
84
| <math>\Delta \boldsymbol{\delta }_{{k}_{c}}=\left( {\bf v}_{j}-{\bf v}_{i}\right) +</math><math>R{\tilde{\bf n}}_{{k}_{c}}\left( {\bf w}_{j}+{\bf w}_{i}\right)</math>
85
|}
86
| style="width: 5px;text-align: right;white-space: nowrap;" | (3)
87
|}
88
89
where <math display="inline">R</math> denotes the particle radius, <math display="inline">{\tilde{\bf n}}_{{k}_{c}}</math> denotes the antisymmetric matrix formed by the components of the unit normal vector <math display="inline">{\bf n}_{{k}_{c}}</math>
90
91
{| class="formulaSCP" style="width: 100%; text-align: center;" 
92
|-
93
| 
94
{| style="text-align: center; margin:auto;" 
95
|-
96
| <math>{\tilde{\bf n}}_{{k}_{c}}=\left[ \begin{matrix}0&{n}_{{k}_{c}3}&-{n}_{{k}_{c}2}\\-{n}_{{k}_{c}3}&0&{n}_{{k}_{c}1}\\{n}_{{k}_{c}2}&-{n}_{{k}_{c}1}&0\end{matrix}\right]</math>
97
|}
98
| style="width: 5px;text-align: right;white-space: nowrap;" | (4)
99
|}
100
101
102
If the contact point <math display="inline">{\bf x}^{{(k}_{c})}</math> is formed by the particles <math>i</math> and <math>j</math>, we have:
103
104
{| class="formulaSCP" style="width: 100%; text-align: center;" 
105
|-
106
| 
107
{| style="text-align: center; margin:auto;" 
108
|-
109
| <math>{\bf n}_{{k}_{c}}=\frac{\left( {\bf x}_{j}-{\bf x}_{i}\right) }{\left| {\bf x}_{j}-{\bf x}_{i}\right| }\quad \quad \quad \, {k}_{c}=</math><math>1,2,\ldots ,{M}_{c}</math>
110
|}
111
| style="width: 5px;text-align: right;white-space: nowrap;" | (5)
112
|}
113
114
115
If the contact point <math display="inline">{\bf x}^{{(k}_{c})}</math> is formed by the particles <math>i</math> and the boundary wall, we have:
116
117
{| class="formulaSCP" style="width: 100%; text-align: center;" 
118
|-
119
| 
120
{| style="text-align: center; margin:auto;" 
121
|-
122
| <math>{\bf n}_{{k}_{c}}=\frac{\left( {x}_{B{k}_{c}}-{x}_{i}\right) }{\left| {x}_{B{k}_{c}}-{x}_{i}\right| }\quad \quad \, {k}_{c}=</math><math>1,2,\ldots ,{M}_{b}</math>
123
|}
124
| style="width: 5px;text-align: right;white-space: nowrap;" | (6)
125
|}
126
127
where <math display="inline">{\bf x}_{B{k}_{c}}</math> denotes the contact point of index <math>k_c</math>  between the particle <math>i</math> and the boundary wall.
128
129
Assuming that the relative displacement <math display="inline">{\Delta \boldsymbol{\delta} }_{{k}_{c}}</math> is small, the change of distance between the two neighbor particles’ centers can be determined as the normal component of their relative displacement, which is set equal to an incremental normal overlap:
130
131
{| class="formulaSCP" style="width: 100%; text-align: center;" 
132
|-
133
| 
134
{| style="text-align: center; margin:auto;" 
135
|-
136
| <math>\begin{matrix}-\Delta {\alpha }_{{k}_{c}}={\bf n}_{{k}_{c}}^{T}{\Delta \boldsymbol{\delta} }_{{k}_{c}}={\bf n}_{{k}_{c}}^{T}\left[ \left( {\bf v}_{j}-{\bf v}_{i}\right) +R{\tilde{\bf n}}_{{k}_{c}}\left( {\bf w}_{j}+{\bf w}_{i}\right) \right] ={\bf n}_{{k}_{c}}^{T}\left( {\bf v}_{j}-{\bf v}_{i}\right) \\\, \Delta {\alpha }_{{k}_{c}}=-{\alpha }_{{k}_{c}}\, ,\, \, if\, \, \Delta {\alpha }_{{k}_{c}}<-{\alpha }_{{k}_{c}}\, \\\, \, {k}_{c}=1,2,\ldots ,{M}_{c}\end{matrix}</math>
137
|}
138
| style="width: 5px;text-align: right;white-space: nowrap;" | (7)
139
|}
140
141
where <math display="inline">\Delta {\alpha }_{{k}_{c}}</math> is positive if the distance between the two centers decreases, while <math display="inline">\Delta {\alpha }_{{k}_{c}}</math> is negative if the distance between the two centers increases. When <math display="inline">\Delta {\alpha }_{{k}_{c}}</math> is negative, <math display="inline">\Delta {\alpha }_{{k}_{c}}</math> can not be smaller than the initial normal overlap <math display="inline">{\alpha }_{{k}_{c}}</math>. <math>M_c</math> is the total number of contacts in the granular media.
142
143
In a similar way, the incremental tangential relative displacement due to the elastic deformations of particles can be written as:
144
145
{| class="formulaSCP" style="width: 100%; text-align: center;" 
146
|-
147
| 
148
{| style="text-align: center; margin:auto;" 
149
|-
150
| <math>\begin{align}{\Delta \boldsymbol{\gamma }}_{{k}_{c}} & =\left( {\bf I}-{\bf n}_{{k}_{c}}{\bf n}_{{k}_{c}}^{T}\right) {\Delta \boldsymbol{\delta} }_{{k}_{c}}\\\, 
151
& =\left( {\bf I}-{\bf n}_{{k}_{c}}{\bf n}_{{k}_{c}}^{T}\right) \left[ \left( {\bf v}_{j}-{\bf v}_{i}\right) +R{\tilde{\bf n}}_{{k}_{c}}\left( {\bf w}_{j}+{\bf w}_{i}\right) \right] \\\, 
152
& =\left( {\bf I}-{\bf n}_{{k}_{c}}{\bf n}_{{k}_{c}}^{T}\right) \left( {\bf v}_{j}-{\bf v}_{i}\right) +R{\tilde{\bf n}}_{{k}_{c}}\left( {\bf w}_{j}+{\bf w}_{i}\right) \quad \quad \quad \, {k}_{c}=1,2,\ldots ,{M}_{c}\end{align}</math>
153
|}
154
| style="width: 5px;text-align: right;white-space: nowrap;" | (8)
155
|}
156
157
158
Further, Eq.(8) can be rewritten as:
159
160
{| class="formulaSCP" style="width: 100%; text-align: center;" 
161
|-
162
| 
163
{| style="text-align: center; margin:auto;" 
164
|-
165
| <math>{\Delta \boldsymbol{\gamma }}_{{k}_{c}}={\tilde{\boldsymbol{\gamma }}}_{{k}_{c}}\Delta {\alpha }_{{k}_{c}}</math>
166
|}
167
| style="width: 5px;text-align: right;white-space: nowrap;" | (9)
168
|}
169
170
where <math display="inline">{\tilde{\boldsymbol{\gamma }}}_{{k}_{c}}</math> denotes the slope vector with respect the unit normal vector <math display="inline">{n}_{{k}_{c}}</math> of the contact elastic relative displacement <math display="inline">{\Delta \boldsymbol{\delta}}_{{k}_{c}}</math>.
171
172
In the same way, when the particle <math>i</math> is in contact with the boundary wall, we have:
173
174
{| class="formulaSCP" style="width: 100%; text-align: center;" 
175
|-
176
| 
177
{| style="text-align: center; margin:auto;" 
178
|-
179
| <math>\begin{matrix}-\Delta \boldsymbol{\alpha }_{{k}_{c}}={\bf n}_{{k}_{c}}^{T}\left( {\overline{\bf v}}_{B{k}_{c}}-{\bf v}_{i}\right) \\\, {\Delta \boldsymbol{\gamma }}_{{k}_{c}}=\, \left( {\bf I}-{\bf n}_{{k}_{c}}{\bf n}_{{k}_{c}}^{T}\right) \left[ {\overline{\bf v}}_{B{k}_{c}}-{\bf v}_{i}+R{\tilde{\bf n}}_{{k}_{c}}{\bf w}_{i}\right] ={\tilde{\boldsymbol{\gamma }}}_{{k}_{c}}\Delta {\alpha }_{{k}_{c}}\quad \, \, {k}_{c}=1,2,\ldots ,{M}_{b}\end{matrix}</math>
180
|}
181
| style="width: 5px;text-align: right;white-space: nowrap;" | (10)
182
|}
183
184
185
The unknown displacements and rotations of the particles are collected in the <math>6N\times 1</math> column matrix <math>\bf U</math>:
186
187
{| class="formulaSCP" style="width: 100%; text-align: center;" 
188
|-
189
| 
190
{| style="text-align: center; margin:auto;" 
191
|-
192
| <math>\begin{align}{\bf U}^{T} & =\left[ \begin{matrix}{\bf V}^{T}&{\bf W}^{T}\end{matrix}\right] \\\, {\bf V}^{T} & ={\left[ {v}_{11}\, \, {v}_{12}\, \, {v}_{13}\, \ldots \, \, {v}_{i1}\, \, {v}_{i2}\, \, {v}_{i3}\, \ldots \quad {v}_{N1}\, \, {v}_{N2}\, \, {v}_{N3}\right] }^{T}\\\, {\bf W}^{T} & ={\left[ {w}_{11}\, \, {w}_{12}\, \, {w}_{13}\, \ldots \, \, {w}_{i1}\, \, {w}_{i2}\, \, {w}_{i3}\, \ldots \quad {w}_{N1}\, \, {w}_{N2}\, \, {w}_{N3}\right] }^{T}\end{align}</math>
193
|}
194
| style="width: 5px;text-align: right;white-space: nowrap;" | (11)
195
|}
196
197
The incremental normal overlaps of contacts between spherical particles are collected in the <math display="inline">M_c \times 1</math> column matrixes  <math>\Delta \boldsymbol{\alpha }_C</math>. The incremental tangential relative displacements are collected in the <math display="inline">3M_c \times 1</math> column matrixes <math>\Delta \boldsymbol{\gamma }_C</math>. In the same way, the incremental relative displacements of the contacts between spherical particles and the boundary wall are collected in the <math display="inline">M_b \times 1</math>  column matrixes <math>\Delta \boldsymbol{\alpha }_B</math>  and in the <math display="inline">3M_b \times 1</math> column matrixes <math>\Delta \boldsymbol{\gamma }_B</math>. Then, Eqs. (3) to (10) can be summarized as:
198
199
{| class="formulaSCP" style="width: 100%; text-align: center;" 
200
|-
201
| 
202
{| style="text-align: center; margin:auto;" 
203
|-
204
| <math>\left\{ \begin{align}\Delta \boldsymbol{\delta} & =\left[ \begin{matrix}-\Delta \alpha \\\Delta \gamma \end{matrix}\right] ={\bf A}^{T}\Delta U\\\, \Delta \boldsymbol{\gamma } & =\tilde{\Upsilon }\Delta \boldsymbol{\alpha } \\\, \left[ \tilde{\Upsilon }\right] & =\left[ {\tilde{\boldsymbol{\gamma }}}_{1}\quad {\tilde{\boldsymbol{\gamma }}}_{2}\, \, \ldots \, \, \ldots \quad {\tilde{\boldsymbol{\gamma }}}_{{M}_{c}+{M}_{b}}\right] \end{align}\right.</math>
205
|}
206
| style="width: 5px;text-align: right;white-space: nowrap;" | (12)
207
|}
208
209
where <math display="inline">{\bf A}^{T}=\left[ \begin{matrix}{\bf n}^{T}&{\bf 0}\\\left( {\bf I}-{\bf nn}^{T}\right) &{\bf R}\tilde{\mathit{\boldsymbol{n}}}\end{matrix}\right]</math>  is a (<math>4M_c + 4M_b)\times 6N</math> matrix composed by unit normal vectors <math display="inline">{\bf n}_{{k}_{c}}</math>.
210
211
===2.2 Statics===
212
213
Due to the deformations of particles, the compressive normal forces and tangential forces are developed. The equilibrium equations of the particles can be written as:
214
215
{| class="formulaSCP" style="width: 100%; text-align: center;" 
216
|-
217
| 
218
{| style="text-align: center; margin:auto;" 
219
|-
220
| <math>\begin{align}{\bf AF}&={\bf A}_{C}{\bf F}_{C}+{\bf A}_{B}{\bf F}_{B}={\bf T}\\\, {\bf F}_{\bf C}^{T}&=\left[ {\bf N}_{C}^{T}\quad \quad {\bf V}_{C}^{T}\right] \\\, {\bf F}_{B}^{T}&=\left[ {\bf N}_{B}^{T}\quad \quad {\bf V}_{B}^{T}\right] \end{align}</math>
221
|}
222
| style="width: 5px;text-align: right;white-space: nowrap;" | (13)
223
|}
224
225
where <math>{\bf N}_B</math> is the <math display="inline">M_b \times 1</math> column matrix of the normal compressive contact forces and <math>{\bf V}_B</math> is the <math display="inline">3M_b \times 1</math> column matrix of the tangential contact forces exerted on the boundary wall; <math>{\bf N}_C</math> is the <math display="inline">M_c \times 1</math> column matrix of the normal compressive contact forces exerted in the contacts between spherical particles and <math>{\bf V}_C</math> is the <math display="inline">3M_c \times 1</math> column matrix of the tangential contact forces; <math>{\bf T}</math> is the <math display="inline">6N\times 1</math>  column matrix of the external forces (such as gravity, boundary forces).
226
227
===2.3 Contact-stiffness model===
228
229
After the equilibrium and kinematic relations are established, a contact force-relative displacement law must be introduced. The DEM simulation in this work is performed using PFC<sup>3D</sup> solver. PFC<sup>3D</sup> provides two stiffness models: a linear model and a simplified Hertz-Mindlin model. In the linear model, the forces and relative displacements are linearly related by the constant contact stiffness. Here, the linear contact model is adopted, which can be written as:
230
231
{| class="formulaSCP" style="width: 100%; text-align: center;" 
232
|-
233
| 
234
{| style="text-align: center; margin:auto;" 
235
|-
236
| <math>\left\{ \begin{matrix}{N}_{{k}_{c}}={k}^{n}{\alpha }_{{k}_{c}}\\\, \Delta {\bf V}_{{k}_{c}}={k}^{s}{\Delta \boldsymbol{\gamma }}_{{k}_{c}}\end{matrix}\right.</math>
237
|}
238
| style="width: 5px;text-align: right;white-space: nowrap;" | (14)
239
|}
240
241
where <math>k^n</math> and <math>k^s</math> are the normal and shear-contact stiffnesses.
242
243
The above relation (14) can be summarized in the form of matrix as follows:
244
245
{| class="formulaSCP" style="width: 100%; text-align: center;" 
246
|-
247
| 
248
{| style="text-align: center; margin:auto;" 
249
|-
250
| <math>\left\{ \begin{matrix}{\bf N} ={\bf K}^{n}\boldsymbol{\alpha} \\\, \Delta {\bf V}={\bf K}^{s}\Delta \boldsymbol{\gamma }\end{matrix}\right.</math>
251
|}
252
| style="width: 5px;text-align: right;white-space: nowrap;" | (15)
253
|}
254
255
256
===2.4 Incremental work===
257
258
We assume that in a given equilibrium state the elastic displacement vector is <math display="inline">\overline{\mathit{\boldsymbol{\delta }}}=</math><math>\left( \overline{\mathit{\boldsymbol{\alpha }}}\, ,\, \overline{\mathit{\boldsymbol{\gamma }}}\right)</math>  and that a small incremental elastic displacement vector <math display="inline">\Delta \boldsymbol{\delta }=</math><math>\left( \Delta \mathit{\boldsymbol{\alpha }}\, ,\, \Delta \mathit{\boldsymbol{\gamma }}\right)</math>  is further applied. Then the incremental work done by the contact forces can be written as:
259
260
{| class="formulaSCP" style="width: 100%; text-align: center;" 
261
|-
262
| 
263
{| style="text-align: center; margin:auto;" 
264
|-
265
| <math>\int_{\overline{\mathit{\boldsymbol{\delta }}}}^{\overline{\mathit{\boldsymbol{\delta }}}+\Delta \delta }{{\bf F}\left( \boldsymbol{\delta } \right) }^{T}d\boldsymbol{\delta } =</math><math>{\overline{\bf F}}^{T}\Delta \delta +\frac{1}{2}\Delta {\bf F}^{T}\Delta \boldsymbol{\delta }</math>
266
|}
267
| style="width: 5px;text-align: right;white-space: nowrap;" | (16)
268
|}
269
270
In the same way, the incremental work done by the external forces can be written as:
271
272
{| class="formulaSCP" style="width: 100%; text-align: center;" 
273
|-
274
| 
275
{| style="text-align: center; margin:auto;" 
276
|-
277
| <math>\int_{\overline{U}}^{\overline{U}+\Delta U}{\bf T}^{T}d{\bf U}={\overline{\bf T}}^{T}\Delta {\bf U}+</math><math>\frac{1}{2}\Delta {\bf T}^{T}\Delta {\bf U}</math>
278
|}
279
| style="width: 5px;text-align: right;white-space: nowrap;" | (17)
280
|}
281
282
283
In view of Eqs. (12) and (13), Eq. (17) can be rewritten as:
284
285
{| class="formulaSCP" style="width: 100%; text-align: center;" 
286
|-
287
| 
288
{| style="text-align: center; margin:auto;" 
289
|-
290
| <math>\int_{\overline{U}}^{\overline{U}+\Delta U}{\bf T}^{T}d{\bf U}={\overline{\boldsymbol{F}}}^{T}{\bf A}^{\mathit{\boldsymbol{T}}}\Delta {\bf U}+</math><math>\frac{1}{2}\Delta {\bf T}^{T}\Delta {\bf U}={\overline{\boldsymbol{F}}}^{T}\Delta \boldsymbol{\delta} +</math><math>\frac{1}{2}\Delta {\bf T}^{T}\Delta {\bf U}</math>
291
|}
292
| style="width: 5px;text-align: right;white-space: nowrap;" | (18)
293
|}
294
295
296
An incremental total energy is defined as:
297
298
{| class="formulaSCP" style="width: 100%; text-align: center;" 
299
|-
300
| 
301
{| style="text-align: center; margin:auto;" 
302
|-
303
| <math>\Psi \left( \Delta {\bf U}\right) =\int_{\overline{\mathit{\boldsymbol{\delta }}}}^{\overline{\mathit{\boldsymbol{\delta }}}+\Delta \delta }{\bf F}\left( \boldsymbol{\delta } \right)^{T}d\boldsymbol{\delta } -</math><math>\int_{\overline{U}}^{\overline{U}+\Delta U}{\bf T}^{T}d{\bf U}=\frac{1}{2}\Delta {\bf F}^{T}\Delta \boldsymbol{\delta } -</math><math>\frac{1}{2}\Delta {\bf T}^{T}\Delta {\bf U}</math>
304
|}
305
| style="width: 5px;text-align: right;white-space: nowrap;" | (19)
306
|}
307
308
===2.5 Minimum principle for incremental total energy===
309
310
In the following, we will assign a superscript * both to any kinematically admissible system of incremental displacements and to any mechanical quantity associated to it by means of Eq. (12) to (19). Then we have:
311
312
{| class="formulaSCP" style="width: 100%; text-align: center;" 
313
|-
314
| 
315
{| style="text-align: center; margin:auto;" 
316
|-
317
| <math>\Psi \left( {\Delta {\bf U}}^{\ast }\right) -</math><math>\Psi \left( \Delta {\bf U}\right) =\frac{1}{2}\left( {\Delta {\bf F}}^{\ast T}{\Delta \boldsymbol{\delta }  }^{\ast }-\right. </math><math>\left. \Delta {\bf F}^{T}{\Delta \boldsymbol{\delta }  }^{\ast }\right) +\frac{1}{2}\Delta {T}^{T}\left( \Delta {\bf U}-\right. </math><math>\left. {\Delta {\bf U}}^{\ast }\right)</math>
318
|}
319
| style="width: 5px;text-align: right;white-space: nowrap;" | (20)
320
|}
321
322
323
In view of Eqs. (12) and (13), Eq. (20) can be rewritten as:
324
325
{| class="formulaSCP" style="width: 100%; text-align: center;" 
326
|-
327
| 
328
{| style="text-align: center; margin:auto;" 
329
|-
330
| <math>\begin{align}\Psi \left( {\Delta {\bf U}}^{\ast }\right) \mathit{\boldsymbol{-}}\Psi \left( \Delta {\bf U}\right) & =\frac{1}{2}\left( {\Delta {\bf F}}^{\ast T}{\Delta \delta }^{\ast }\mathit{\boldsymbol{-}}\Delta {\bf F}^{T}{\Delta \boldsymbol{\delta }  }^{\ast }\right) \mathit{\boldsymbol{+}}\frac{1}{2}{\Delta {\bf F}}^{T}{A}^{\mathit{\boldsymbol{T}}}\left( \Delta U\mathit{\boldsymbol{-}}{\Delta {\bf U}}^{\ast }\right) \\\, & =\frac{1}{2}\left( {\Delta {\bf F}}^{\ast T}\mathit{\boldsymbol{-}}\Delta {\bf F}^{T}\right) {\Delta \boldsymbol{\delta }  }^{\ast }\mathit{\boldsymbol{+}}\frac{1}{2}{\Delta {\bf F}}^{T}\left( \Delta \boldsymbol{\delta }  \mathit{\boldsymbol{-}}{\Delta \boldsymbol{\delta }  }^{\ast }\right) \\\, & =\frac{1}{2}{\left( {\Delta \boldsymbol{\delta }  }^{\ast }\mathit{\boldsymbol{-}}\Delta \boldsymbol{\delta }  \right) }^{\mathit{\boldsymbol{T}}}{\bf K}\left( {\Delta \boldsymbol{\delta }  }^{\ast }\mathit{\boldsymbol{-}}\Delta \boldsymbol{\delta } \right) \end{align}</math>
331
|}
332
| style="width: 5px;text-align: right;white-space: nowrap;" | (21)
333
|}
334
335
336
where <math display="inline">K=\left[ \begin{matrix}{\bf K}^{n}&{\bf 0}\\{\bf 0}&{\bf K}^{s}\end{matrix}\right]</math>  is the stiffness matrix, which is diagonal and definite positive. it can be noted that, <math display="inline">\Psi \left( {\Delta {\bf U}}^{\ast }\right) -</math><math>\Psi \left( \Delta {\bf U}\right)</math>  is always positive. So it can be concluded that, The incremental total energy <math display="inline">\Psi \left( {\Delta {\bf U}}^{\ast }\right)</math>  attains its minimum if the kinematically admissible system of incremental displacements is such to satisfy the equilibrium equations (13).
337
338
==3. Numerical model==
339
340
The minimum principle for incremental total energy <math display="inline">\Psi \left( {\Delta U}^{\ast }\right)</math>  proposed in the section 2, can be used to accelerate the convergence of DEM simulation. The numerical form of the minimum principle can be written as:
341
342
{| class="formulaSCP" style="width: 100%; text-align: center;" 
343
|-
344
| 
345
{| style="text-align: center; margin:auto;" 
346
|-
347
| <math>\begin{matrix}min\quad \frac{1}{2}{\Delta {U}^{\mathit{\boldsymbol{\ast }}}}^{T}\mathit{\boldsymbol{\cdot }}AK{A}^{T}\cdot \Delta {U}^{\mathit{\boldsymbol{\ast }}}-\frac{1}{2}\Delta {T}^{T}\Delta {U}^{\mathit{\boldsymbol{\ast }}}\quad \\\, s.t.\, \, \left\{ \begin{matrix}K{A}_{\mathit{\boldsymbol{B}}}^{\mathit{\boldsymbol{T}}}\Delta {U}^{\mathit{\boldsymbol{\ast }}}=\Delta {F}_{\mathit{\boldsymbol{B}}}\\\, {A}^{T}\mathit{\boldsymbol{=}}\left[ \begin{matrix}{n}^{T}&0\\\left( I-n{n}^{T}\right) &R\tilde{\mathit{\boldsymbol{n}}}\end{matrix}\right] \\\, K\mathit{\boldsymbol{=}}\left[ \begin{matrix}{K}^{n}&0\\0&{K}^{s}\end{matrix}\right] \end{matrix}\right. \\\, \end{matrix}</math>
348
|}
349
| style="width: 5px;text-align: right;white-space: nowrap;" | (22)
350
|}
351
352
where <math>{\bf F}_B</math> denotes the contact forces exerted on the boundary wall, which is equal to the external boundary force.
353
354
In this work, the movement and interaction of spherical particles is modeled by PFC<sup>3D</sup>. In the later stage of PFC<sup>3D</sup> modeling, when the new contacts are no longer detected and created, and the old contacts are no longer broken and deleted, Eq. (22) is introduced in order to reach mechanical equilibrium quickly. The improved DEM formulation is shown in [[#img-3|Figure 3]].
355
356
<div id='img-3'></div>
357
{| class="wikitable" style="margin: 0em auto 0.1em auto;border-collapse: collapse;width:auto;" 
358
|-style="background:white;"
359
|style="text-align: center;padding:10px;"| [[Image:Review_294821678474-image3.png|600px]]
360
|-
361
| style="background:#efefef;text-align:left;padding:10px;font-size: 85%;"| '''Figure 3'''. Improved DEM formulation based on the minimum principle for incremental total energy
362
|}
363
364
365
Generally the tangential contact force is smaller or much smaller than the corresponding normal contact forces. Especially in the later stage of DEM simulation, the particles are unbalanced mainly subject to the normal contact forces. In order to require modest sizes of computer memory, considering the incremental work is mainly done by the normal contact forces, the numerical srategy can be rewritten as:
366
367
{| class="formulaSCP" style="width: 100%; text-align: center;" 
368
|-
369
| 
370
{| style="text-align: center; margin:auto;" 
371
|-
372
| <math>\begin{matrix}min\quad \frac{1}{2}{\Delta {V}^{\mathit{\boldsymbol{\ast }}}}^{T}\mathit{\boldsymbol{\cdot }}A{K}^{n}{A}^{T}\cdot \Delta {V}^{\mathit{\boldsymbol{\ast }}}-\frac{1}{2}\Delta {T}^{T}\Delta {V}^{\mathit{\boldsymbol{\ast }}}\quad \\\, s.t.\, \, \left\{ \begin{matrix}{A}^{T}\mathit{\boldsymbol{=}}{n}^{T}\\\, {K}^{n}{A}_{\mathit{\boldsymbol{B}}}^{\mathit{\boldsymbol{T}}}\Delta {V}^{\mathit{\boldsymbol{\ast }}}=\Delta {F}_{\mathit{\boldsymbol{B}}}\end{matrix}\right. \\\, \end{matrix}</math>
373
|}
374
| style="width: 5px;text-align: right;white-space: nowrap;" | (23)
375
|}
376
377
378
Eq. (23) is also introduced into the improved DEM formulation shown in [[#img-3|Figure 3]] to observe the computational efficiency.
379
380
The numerical calculation of the minimum principle is performed on MATLAB. The MATLAB provides a solver “quadprog” for quadratic objective functions with linear constraints. The solver “quadprog” finds a minimum for a problem specified by:
381
382
{| class="formulaSCP" style="width: 100%; text-align: center;" 
383
|-
384
| 
385
{| style="text-align: center; margin:auto;" 
386
|-
387
| <math>\begin{matrix}\mathrm{\underbrace{min}_{x}}\,\frac{1}{2}{x}^{T}Hx+{f}^{T}x\quad \\\, s.t.\, \, \left\{ \begin{matrix}A\cdot x\leq b\\Aeq\cdot x=beq\\lb\leq x\leq ub\end{matrix}\right. \\\, \end{matrix}</math>
388
|}
389
| style="width: 5px;text-align: right;white-space: nowrap;" | (24)
390
|}
391
392
where <math>H</math>, <math>A</math>, and <math>Aeq</math> are matrices, and <math>f</math>, <math>b</math>, <math>beq</math>, <math>lb</math>, <math>ub</math>, and <math>x</math> are vectors. The solver “quadprog” in MATLAB can be used to solve Eq. (22) or (23).
393
394
The improved DEM formulation completed after a PFC<sup>3D</sup> final check, because of the FEM-DEM coupling settings and the introduction of Eq. (23) that the tangential contact force is neglected.
395
396
==4. Examination of the numerical model==
397
398
The improved DEM formulation is examined numerically using the 3D FEM-DEM coupling model reported by Liu et al. [12]. The units of length/mm, force/N, time/s, stress/MPa are used in the coupling model.
399
400
In the FEM model, an elbow tube of stainless steel (1Cr18Ni9Ti) with wall thickness <math>t=0.3</math>mm, outer diameter <math>D=30</math>mm, bending radius <math>R=90</math>mm, is formed. The material parameters of 1Cr18Ni9Ti stainless steel are listed in [[#tab-1|Table 1]], and the stress-strain curve is shown in [[#img-4|Figure 4]].
401
402
<div class="center" style="font-size: 75%;">'''Table 1'''. Room temperature mechanical properties of 1Cr18Ni9Ti </div>
403
404
<div id='tab-1'></div>
405
{| class="wikitable" style="margin: 1em auto 0.1em auto;border-collapse: collapse;font-size:85%;width:auto;" 
406
|-style="text-align:center"
407
! Young's modulus (GPa) !! Poisson ratio !! Density (kg/m<sup>3</sup>) !! Yield strength (MPa)!! Tensile strength (MPa)!! Elongation (%)
408
|-style="text-align:center"
409
|  202
410
|  0.29
411
|  7850
412
|  219
413
|  755
414
|  48
415
|}
416
417
418
<div id='img-4'></div>
419
{| class="wikitable" style="margin: 0em auto 0.1em auto;border-collapse: collapse;width:auto;" 
420
|-style="background:white;"
421
|style="text-align: center;padding:10px;"| [[Image:Review_294821678474-image4-c.jpeg|336px]]
422
|-
423
| style="background:#efefef;text-align:left;padding:10px;font-size: 85%;"| '''Figure 4'''. Stress-strain curve of 1Cr18Ni9Ti
424
|}
425
426
427
The FEM model of ABAQUS [13] shown in [[#img-5|Figure 5]](a), considers nonlinear geometry through a large-strain description of elbow tube, and material is considered both elastic and elastic-plastic (1Cr18Ni9Ti stainless steel). The FEM model is three-dimensional, and four-node reduced-integration shell elements (S4R) are employed for modeling of the thin-wall elbow tube. Contact forces between particles and tube are imposed on the four nodes nearby the particle-tube contact points in the form of concentrated force. Regarding the number of elements in the longitudinal direction, at least 5 elements per particle size are employed in the tube. Friction coefficient between die and tube is set 0.05.
428
429
<div id='img-5'></div>
430
{| class="wikitable" style="margin: 0em auto 0.1em auto;border-collapse: collapse;width:auto;" 
431
|-style="background:white;"
432
|style="text-align: center;padding:10px;"| [[Image:Review_294821678474-image5.png|516px]]
433
|-
434
| style="background:#efefef;text-align:left;padding:10px;font-size: 85%;"| '''Figure 5'''. FEM-DEM coupling model of the elbow tube push-bending. (a) FEM model. (b) DEM model
435
|}
436
437
438
It is shown in [[#img-5|Figure 5]](b) that, the cast iron granular media with particle diameter 1.58mm is modeled as filler by DEM. The granulometric distribution of granular media is uniform (monodisperse). Wall boundary is created according to the tube shape determined in FEM. The linear model with constant stiffness is employed to model the particle-wall and inter-particle contacts. The friction coefficients of both particle-wall and inter-particle are set 0.05. The properties of cast iron granular media are listed in [[#tab-2|Table 2]].
439
440
<div class="center" style="font-size: 75%;">'''Table 2'''. Mechanical properties of cast iron granular media</div>
441
442
<div id='tab-1'></div>
443
{| class="wikitable" style="margin: 1em auto 0.1em auto;border-collapse: collapse;font-size:85%;width:auto;" 
444
|-style="text-align:center"
445
!  Young's modulus (GPa) !! Shear modulus (GPa) !! Poisson's ratio !! Density (kg/m<sup>3</sup>)
446
|-
447
|  style="text-align: center;"|150
448
|  style="text-align: center;"|58
449
|  style="text-align: center;"|0.3
450
|  style="text-align: center;"|7300
451
|}
452
453
454
In the FEM-DEM coupling program, at the <math display="inline"> i </math>-th increment, the shape of tube is calculated in FEM under the force boundary conditions (contact forces between particles and tube) determined at the (<math display="inline">i-1</math>)th increment in DEM. The contact forces distribution in granular media is calculated in DEM under the displacement boundary conditions (tube shape) determined at the (<math display="inline">i-1</math>)th increment in FEM.
455
456
The PFC<sup>3D</sup>, ABAQUS and MATLAB use different default computational accuracy to complete the simulation or optimization. The default computational accuracy of the solver “quadprog” in MATLAB (tolerance <math display="inline">\le 10^{-6}</math>) is much smaller than that of PFC<sup>3D </sup>(the maximum unbalanced force <math display="inline">\le 10^{-2}</math>). For comparison, the completed condition of PFC<sup>3D</sup> simulation is set to “the maximum unbalanced force <math display="inline">\le 10^{-6}</math>”.
457
458
The time-consuming results of the different FEM-DEM coupling models with no minimum principle, with Eq. (22) and with Eq. (23) under the same simulation condition, are listed in [[#tab-3|Table 3]].
459
460
<div class="center" style="font-size: 75%;">'''Table 3'''. The time-consuming results of the different FEM-DEM coupling models</div>
461
462
<div id='tab-1'></div>
463
{| class="wikitable" style="margin: 1em auto 0.1em auto;border-collapse: collapse;font-size:85%;width:auto;" 
464
|-style="text-align:center"
465
! [[Image:Review_294821678474-image6.png|122px]] !! Not using the minimum principle !! Using Eq. (22) !! Using Eq. (23)
466
|-
467
|  style="text-align: center;"|Time consuming (hour)
468
|  style="text-align: center;"|≈210
469
|  style="text-align: center;"|≈170
470
|  style="text-align: center;"|≈120
471
|}
472
473
474
It is shown in [[#tab-3|Table 3]] that, the improved FEM-DEM coupling simulation using the minimum principle Eq. (22) or Eq. (23) is less time-consuming than the previous coupling model.
475
476
==5. Conclusions==
477
478
Considering the new contacts are no longer created and the old contacts are no longer deleted in the later stage of DEM simulation, an improved numerical formulation is proposed in order to reach mechanical equilibrium quickly and accelerate the convergence of DEM simulation. The improved numerical formulation is based on the minimum principle for incremental total energy in granular media. The improved DEM formulation is examined numerically using the 3D FEM-DEM coupling model of the granular-media-based thin-wall elbow tube push-bending process. It is demonstrated that the improved FEM-DEM coupling model is less time-consuming than the previous under the same simulation condition.
479
480
==Funding Information==
481
482
The present work is funded by the Scientific Research Foundation of SIIT (2022kyqd001), Natural Science Research General Project of Universities in Jiangsu Province (no.21KJB430006).
483
484
==References==
485
486
<div class="auto" style="text-align: left;width: auto; margin-left: auto; margin-right: auto;font-size: 85%;">
487
488
[1] Yang H.,  Li H.,  Zhang Z.Y., Zhan M.,  Liu J.,  Li G. Advances and trends on tube bending forming technologies. Chinese J. Aeronaut.,  25(1):1-12, 2012.
489
490
[2]  Teng B.G.,  Hu L.,  Liu G.,  Yuan S.J. Wrinkling behavior of hydro bending of carbon steel Al alloy bilayered tubes. Trans. Nonferrous Met. Soc. China, 22:560−565, 2012.
491
492
[3] Yang H.,  Li H.,  Zhang Z.Y.,  Zhan M.,  Liu J.,  Li G. Advances and trends on tube bending forming technologies. J. Plast. Eng., 8:83–85, 2012.
493
494
[4]  Zeng Y.,  Li Z. Experimental research on the tube push-bending process. Journal of Materials Processing Technology, 122(2-3):237-240, 2002.
495
496
[5] Oh I.Y., Han S.W., Woo Y.Y., Ra J.H., Moon Y.H. Tubular blank design to fabricate an elbow tube by a push-bending process. Journal of Materials Processing Technology, 260:112-122, 2018.
497
498
[6] Liu H.,  Zhang S.H.,  Ding Y.P., Shi G.L., Geng Z. A simplified formulation for predicting wrinkling of thin-wall elbow tube in granular media–based push-bending process. International Journal of Advanced Manufacturing Technology, 115:541-549, 2021.
499
500
[7]  Du B.,  Zhao C.C.,  Dong G.J., Bi J. FEM-DEM coupling analysis for solid granule medium forming new technology. J. Mater. Process. Technol., 249:108-117, 2017.
501
502
[8]  Chen H.,  Güner A.,  Khalifa N.B.,  Tekkaya A.E. Granular media-based tube press hardening. J. Mater. Process. Technol., 228:145–159, 2016.
503
504
[9]  Chen H.,  Hess S.,  Haeberle J.,  Pitikaris S.,  Born P.,  Güner A.,  Sperl M.,  Tekkaya A.E.  Enhanced granular medium-based tube and hollow profile press hardening. CIRP Ann.-Manuf. Technol., 65:273–276, 2016.
505
506
[10]  Dong G.J.,  Zhao C.C.,  Peng Y.X.,  Li Y. Hot granules medium pressure forming process of AA7075 conical parts. Chin. J. Mech. Eng., 28:580–591, 2015.
507
508
[11]  Dong G.J.,  Bi J.,  Du B.,  Zhao C.C. Research on AA6061 tubular components prepared by combined technology of heat treatment and internal high pressure forming. J. Mater. Process. Technol., 242:126–138, 2017.
509
510
[12]  Liu H.,  Zhang S.H.,  Song H.W., Shi G.L., Cheng M. 3D FEM-DEM coupling analysis for ranular-media-based thin-wall elbow tube push-bending process. International Journal of Material Forming, 12:985-994, 2019.
511
512
[13]  Hibbit H.D., Karlsson B.I.,  Sorensen P. Theory Manual, ABAQUS, Version 6.7. Providence, RI, USA, 2007.
513
514
<!--
515
----
516
517
E-mail address: hliu[mailto:@ @]<span style="text-align: center; font-size: 75%;">alum.imr.ac.cn</span>
518

Return to Liu et al 2023a.

Back to Top

Document information

Published on 30/05/23
Accepted on 19/05/23
Submitted on 08/05/23

Volume 39, Issue 2, 2023
DOI: 10.23967/j.rimni.2023.05.005
Licence: CC BY-NC-SA license

Document Score

0

Views 33
Recommendations 0

Share this document

claim authorship

Are you one of the authors of this document?