产品
0
00:00:00,100 --> 00:00:02,080
大家早上 我叫史蒂夫·黑尔
Morning everyone. My name is Steve Hale.
1
00:00:02,240 --> 00:00:02,750
我是一个高级
I'm a senior
2
00:00:03,080 --> 00:00:04,440
我们公司的工程经理
engineering manager here at c. A.
3
00:00:04,600 --> 00:00:11,200
同事 今天我要讲的是如何使非线性分析发散
Associates. Today I'm going to talk about how we can get our nonlinear analysis to diverge.
4
00:00:11,360 --> 00:00:18,850
这是一个很常见的问题 特别是对于那些 首先从非线性f开始的人
And this is a very common problem, uh, especially for people who, 1st starting out with, uh, running nonlinear f.
5
00:00:18,910 --> 00:00:21,550
b A 呃 问题 嗯
b. A. UH, problems. Um,
6
00:00:22,630 --> 00:00:29,030
演示将持续大约30到40分钟 所有人都静音
the presentation will last, uh, approximately 30 to 40 min UH, everyone is muted.
7
00:00:29,070 --> 00:00:35,940
呃 如果你们有任何问题 只要在聊天窗口中输入 我会在演示结束时解决
UH. So if you do have any questions, just type them into the chat window, and I'll try to address those at the end of the presentation.
8
00:00:36,940 --> 00:00:42,380
好的 嗯 我只想简单介绍一下我们公司的背景
Okay, uh. I just want to give a little background on our company.
9
00:00:42,510 --> 00:00:45,250
呃 我们是c A Associates
UH. We are c. A. Associates.
10
00:00:45,410 --> 00:00:52,310
我们是康涅狄格州明德伯里的一家工程咨询公司 我们的专长是 结构和安全分析
We're an engineering consulting firm here in Middlebury, connecticut, and we specialize in both, uh, structural and safety analysis.
11
00:00:52,580 --> 00:00:55,760
我们从1985年开始就是混蛋的渠道合作伙伴
We've been an ass' channel partner since 1985.
12
00:00:57,600 --> 00:01:14,340
这个演讲是我们蚂蚁式电子学习系列的一部分 我们大约每个月做一次 关于各种不同的主题 我们相信这将对奥克兰大学的答案 答案社区 奥克兰大学和我们的客户有帮助
Um, This presentation is part of our ants of e learning series, which we give about once every month, on a variety of different topics that we believe will be helpful for UH answers, the answers community, UM and our customers.
13
00:01:15,390 --> 00:01:19,730
UM我们的网站是c
UM In terms of our website, it's c.
14
00:01:19,890 --> 00:01:21,750
a e a i com
a. e. a. i. dot com.
15
00:01:22,040 --> 00:01:29,640
我知道这有点拗口 但是 如果你去那个网站 那里有很多资源
UH, I know that's a mouthful, but, um, if you go to that site, there are quite a few resources there.
16
00:01:29,690 --> 00:01:33,250
嗯 那可能会很有帮助
UM that, uh, that could be quite helpful.
17
00:01:33,630 --> 00:01:35,430
其中之一就是我们的资源库
One of them is our resource library.
18
00:01:35,590 --> 00:01:43,450
它有 UM 它有更新研讨会的链接 我们在过去每次修改答案时都会举办研讨会
It has, UM, it has links to update seminars that we've given in the past for each new revision of answers.
19
00:01:43,610 --> 00:01:53,440
嗯案例研究 我们写了很多关于我们做过的不同咨询项目的不同案例研究 还有各种各样的回答报告
Um case studies. We've written quite a few different case studies on different consulting projects that we've done, and, uh, variety of answers presentations.
20
00:01:53,600 --> 00:02:10,730
另一个很好的资源是我们的工程优势博客 你们可以从我们的网站上访问 我们每周发布一次 内容涉及很多不同的 UH主题 从c
Another excellent resources resource is our engineering advantage blog um, which you can access from our website, and we put those out once a week, on again, on a variety of different, UH topics, um ranging from c.
21
00:02:10,830 --> 00:02:17,380
从f D 到结构分析 有些是很普遍的有些是具体的
f. D. to structural analysis. UH, some of them very general and some of them specific the um.
22
00:02:17,500 --> 00:02:25,580
我们还有一个更简单的培训重新注册日历 呃 所以你可以在那里注册培训
We also have an easier training re- registration calendar, UH, so you can register for training there.
23
00:02:25,740 --> 00:02:38,520
你可以看看芝加哥大学的链接 它会给你一些指导 关于你可以上什么课程 呃 呃 最适合你的申请
And there's a c. A. university link that you can look into, and that will give you some guidance on what classes, UH, you could take to, um, uh, that are best for your application.
24
00:02:40,040 --> 00:02:58,430
是的 好的 嗯 反对ELearning的演讲都被记录下来了 你可以直接从我们的网站UH访问 或者你可以访问我们的youtube频道 那里有我们的youtube频道链接
Yeah, okay, um. The antis E Learning UM presentations have all been recorded, and you can access those either directly from our website, UH, or you can go to our youtube channel, and there's the link to our UH youtube channel right there.
25
00:03:00,000 --> 00:03:05,200
这个也会被录下来 并很快发布出去
And this one will be recorded and posted, uh, very soon as well.
26
00:03:06,350 --> 00:03:11,950
好了 我们继续 非线性收敛的话题
Okay, let's move on to to the, uh, nonlinear convergence topics.
27
00:03:11,950 --> 00:03:19,440
首先 我要讲非线性分析 什么是收敛
Um 1st, I'm gonna talk about nonlinear analysis, what it is, and, um, what is convergence?
28
00:03:19,800 --> 00:03:25,660
是什么导致了收敛 我们如何获得收敛性 以及不收敛的主要原因
Uh, what causes convergence? How we obtain convergence, and the primary causes of non convergence.
29
00:03:25,820 --> 00:03:28,420
然后我将演示一个问题
Then I'll go through through a demonstration problem.
30
00:03:30,060 --> 00:03:35,100
非线性分析有三种主要类型
There are three major types of of non linear analysis.
31
00:03:35,390 --> 00:03:38,630
第一个是几何非线性
1st one is geometric non linearities.
32
00:03:38,790 --> 00:03:43,130
这个包含了 三个非线性子类
And this one really includes, uh, three sub classes non linearities.
33
00:03:43,560 --> 00:03:51,500
嗯 大变形 有些人称之为大旋转 也叫大应变和应力硬化
Uh, large deformations. Uh, some people call it large rotations, also large strains and stress stiffening.
34
00:03:51,930 --> 00:03:58,530
这里发生的是 嗯 负荷矢量的方向
What happens here is, um, the uh, the direction of the load vector.
35
00:03:58,960 --> 00:04:01,100
在整个分析过程中会发生变化
UH, can change throughout the analysis.
36
00:04:01,140 --> 00:04:06,760
在分析过程中 最刚性矩阵需要更新
The stiffest matrices need to be updated, UH, during the analysis.
37
00:04:06,840 --> 00:04:13,880
同样 作为组成的元素 那些刚度矩阵必须被更新
UM Also, as the elements to form, those stiffness matrices have to be updated.
38
00:04:14,030 --> 00:04:21,910
UM根据几何优势的类型 你可以得到加强和软化响应
UM You can get both a stiffening and a softening response depending on the type of geometric dominarity.
39
00:04:21,990 --> 00:04:24,990
那是在分析中发生的
Uh, that that's occurring in the analysis.
40
00:04:25,070 --> 00:04:30,310
现在 好消息是 很容易激活这些非线性效应
Now, the good news there is that it's very easy to activate those nonlinear effects.
41
00:04:30,750 --> 00:04:34,210
第二种是材料非线性
Um, the the 2nd one is material nonlinearities.
42
00:04:34,370 --> 00:04:37,610
有许多不同类型的材料非线性-非线性
And there are a number of different types of material non le- non linearities.
43
00:04:37,770 --> 00:04:41,330
最常见的可能是金属的可塑性
Probably the most common one is is metal plasticity.
44
00:04:41,670 --> 00:04:49,340
还有一些其他的 比如 粘塑性 粘弹性 蠕变 这是很常见的
UH, but some of the other ones you can have are, uh, say, visco plasticity, viscolasticity, creep as a co- as is not an uncommon one.
45
00:04:49,560 --> 00:05:01,030
还有超塑性 用于橡胶之类的材料 还有一些塑料 比如金属塑性 你会有一个非线性响应 因为
And hyperlasticity, which is used for, uh, rubber like materials, and some plastics as well, in metal plasticity, e.g., you have a nonlinear response, because
46
00:05:01,280 --> 00:05:05,710
一旦达到屈服点 材料就会软化
once you reach that yield point, the material softens and and that
47
00:05:06,040 --> 00:05:14,270
曲线 你们在右边看到的图表显示了一个非常简单的压力排名
curve. That um graph that you see on the right shows the stress ranker for just a very simple um
48
00:05:14,580 --> 00:05:24,010
嗯金属 最后一个可能是最难解的 那就是接触非线性
um metal. The last one is probably the most difficult one to solve for, and that's a contact nonlinearity um.
49
00:05:24,310 --> 00:05:25,030
这也是
It's also
50
00:05:25,190 --> 00:05:28,380
称为变化状态型非线性
known as a changing status type nonlinearity um.
51
00:05:28,640 --> 00:05:30,560
f e
And the way that f. e.
52
00:05:30,660 --> 00:05:48,600
a 代码代码处理这是当两个部分相互接触时 它会激活额外的刚度这是一个非常高的刚度在最硬的矩阵中突然被激活 这在很多情况下会导致一些严重的收敛问题
A. codes codes handle this is when two parts come into contact with each other, it activates additional stiffness um and that's a very high stiffness that suddenly gets activated in the uh, in the stiffest matrix, and that can cause some serious convergence problems in many cases.
53
00:05:50,600 --> 00:05:56,620
现在 answers使用牛顿raps算法来获得一个收敛解
Now, answers uses a Newton raps An algorithm to obtain a converge solution.
54
00:05:56,780 --> 00:06:05,530
它所做的就是平衡结构内部的内力和外加的外部载荷
So what it's doing here is trying to balance the internal forces inside the structure with the applied external load.
55
00:06:05,830 --> 00:06:21,720
嗯 因为这些方程是用UM来解的 用线性方程 UM 它在解的时候要不断地更新切线和刚度的距离矩阵
UM. And because the the equations are solved using, UM, using, uh uh, linear equations, UM, it has to constantly update the tangent, stiffness made distance matrix as it's solving.
56
00:06:21,960 --> 00:06:29,690
所以它必须经过一系列的迭代 你可以在右边的图表中看到
So it has to go through a series of iterations, and you'll see this in the graph on the right.
57
00:06:30,110 --> 00:06:34,470
UM在解的过程中 不断更新k
UM as it solves, it's constantly updating that k.
58
00:06:34,630 --> 00:06:42,970
T 刚度矩阵 直到施加的力与内力平衡
T. stiffness matrix until the applied forces are balanced by the internal forces.
59
00:06:43,130 --> 00:06:45,550
这就是fnr向量
That's the f. n. R. vector.
60
00:06:46,210 --> 00:07:02,460
所以 这些迭代还在继续 一般来说 如果你有一个非常平滑的强制偏转响应 那么牛顿算法就做得很好 它可以处理它而不需要太多的修改
So, um, those iterations continue. Now, typically, if you have a very smooth forced deflection response, then the Newton raps and algorithm does a very good job, and it can handle it without much in the way of modifications.
61
00:07:02,500 --> 00:07:13,660
然而 当你有接触 或者如果你有一个软化响应 然后是一个加强响应 它可能会变得相当困难 得到一个收敛解
However, when you have contact, or if you have a softening followed by a stiffening response, it could become quite difficult to obtain a converge solution.
62
00:07:16,540 --> 00:07:20,660
我们来看看不收敛的主要原因
So let's look at the primary causes of non convergence.
63
00:07:21,290 --> 00:07:23,990
呃 第一种是刚度过低
Uh. The 1st one is excessively low stiffness.
64
00:07:24,150 --> 00:07:37,660
如果你有一个零件的整个部分 非常大的一部分 那就变成了塑料 它进入塑料状态 刚度 呃 项变得很低 呃 材料变得很软
If you have a full section of a part, a very large section of part, that becomes plastic, it goes to the plastic regime, the stiffness, uh, terms become very low, UH, the material becomes very soft.
65
00:07:37,700 --> 00:07:54,630
然后 在刚度上的巨大变化会导致收敛 问题 屈曲或其他结构不稳定 UM uh 压缩应力 会导致 导致结构的整体刚度急剧下降
And then, uh, that dramatic change in stiffness causes convergence, problems, buckling or other structural instability that, UM, that UH, compressive stress, causes the, causes the overall stiffness of the structure to decrease dramatically.
66
00:07:55,510 --> 00:08:02,800
UM如果这种僵硬突然改变 就会引起问题 身体就会突然相互接触
UM If this stiffness changes abruptly, that cause problems, the body suddenly come into contact with each other.
67
00:08:03,330 --> 00:08:23,380
嗯如果上下文 如果这是太高了 嗯 通常发生在你有灵活的结构 就像如果你有梁cantally掉 或者如果你有 呃 板壳 嗯 嗯松散和不受约束的 呃 他们接触到另一个部分
UH if the context, if this is too high, UM, that typically occurs when you have flexible structures, like if you have beams that are cantally rid out, or if you have, uh, plate shells that are, um, that are, that are, uh loose and unconstrained, UH, they come into contact with another part.
68
00:08:23,400 --> 00:08:38,290
发生的事情是 当它们接触时 接触力可以有效地把它们推离表面 然后 内部应力会导致那个部分反弹回来 再次撞击表面
UH, what happens there is that when they come into contact, UH, that contact force can effectively push them off the surface, then the, uh, internal stresses can cause that part to then bounce back and hit the surface again.
69
00:08:38,540 --> 00:08:40,920
这就是接触抖振类型的问题
And that the contact chattering type of problem.
70
00:08:41,080 --> 00:08:47,710
呃 你会看到力的收敛 啊 历史在这种情况下振荡
Uh? And you you, you will see the force convergence, ah, history oscillating that case.
71
00:08:48,550 --> 00:08:55,350
严重的元素扭曲 这是另一个常见的问题 当你在尖锐的角落接触时会发生
Um, severe element distortion. That's another common problem, uh, that can occur when you have contact at sharp corners.
72
00:08:55,510 --> 00:09:00,290
所以把这些角弄圆是很有帮助的 加入一些软下疳或混合物
So it can be helpful to round out those corners, put in, uh, chanceres or blends.
73
00:09:00,350 --> 00:09:13,480
或者 如果你之前有极端非线性的材料变形 嗯 这些可能是由可塑性或超弹性引起的 例如 嗯 脊体运动
Uh, or if you ex- have extreme nonlinear material deformation, uh, those can be caused by, uh, plasticity or hyperlasticity, e.g., um, ridge body motion.
74
00:09:13,630 --> 00:09:24,560
另一个典型的问题 当你进行接触分析时 通常会遇到这种情况 接触和身体之间有间隙 没有足够的约束
Another typical problem. UH, usually you run into that when you have a contact analysis, and you have a gap between contact and body, that there aren't enough constraints applied.
75
00:09:24,880 --> 00:09:27,680
UM这可能会引起问题
UM That can cause an issue.
76
00:09:27,860 --> 00:09:33,080
再一次接触到尸体 那是开放式的 可能会造成问题
Uh, contact bo- contacted the bodies again, that start off in open configuration, that can cause a problem.
77
00:09:33,490 --> 00:09:44,550
还有 如果你在一个洞里有一个大头针 你没有任何摩擦 大头针就可以在那个洞里旋转 或者叫波洞
Um, also, if you have, say, a, uh, a pin in a hole, and you don't have any friction, that pin can then spin in that hole, or bolton hole.
78
00:09:45,390 --> 00:09:52,540
所以 呃 那是它可以是一个非线性 呃 刚体运动问题
So that, uh, that's It could be a non linear, um, uh, rigid body motion problem.
79
00:09:52,860 --> 00:09:58,060
好吧 在我们进入演示之前 让我问你一个问题
All right, let me ask you a question here before we go into the demo.
80
00:09:58,140 --> 00:10:02,970
什么模型属性最容易引起收敛问题
Um, what model property causes you the most convergence problems?
81
00:10:03,130 --> 00:10:04,510
contactlasticity
Is it contactlasticity,
82
00:10:04,510 --> 00:10:14,400
蠕变 屈曲 大变形和张力 或者其他 导致你的通话者出现问题的行为
creep, buckling, large deformations and strains, or some other type of, uh, behavior that's causing your conversers problems?
83
00:10:14,620 --> 00:10:19,190
他们只会给你大约一分钟的时间来思考和回答这个问题
They'll just give you about, uh, about a minute to think about that and answer that question,
84
00:10:19,580 --> 00:10:24,110
然后 然后我们再进行演示
and then, uh, then we'll move on to the demonstration.
85
00:10:47,080 --> 00:10:49,920
好的 希望每个人都有机会回答这个问题
Okay, hopefully everybody's had a chance to answer that.
86
00:10:52,640 --> 00:10:54,880
让我们快速看一下结果
Let's take a quick look at the results.
87
00:10:57,400 --> 00:11:11,100
好了 在我看来 除了少数例外 你们大多数人都认为接触是一种成熟的 模型属性导致了大多数收敛问题
Okay, so it looks to me, with a few exceptions, the majority of you have responded that contact is the matur-, the model property that causes you the most convergence problems.
88
00:11:11,160 --> 00:11:18,150
这是可以理解的 因为这是一个高度非线性的行为
And that's that's understandable, because that is a highly non linear uh behavior.
89
00:11:18,950 --> 00:11:25,800
嗯 有滑动的问题 他们的问题是 最初的接触间隙
Um. There are issues with sliding, their issues with, uh, initial contact gaps.
90
00:11:25,840 --> 00:11:30,560
当我们讲到演示问题时 我们会详细讨论这个问题
Um And we'll talk about this more when we get into the demonstration problem.
91
00:11:30,870 --> 00:11:35,610
澳大 因为这个问题包括平面 塑性和接触
UM Because that problem includes both pla-, both plasticity and contact.
92
00:11:41,610 --> 00:11:46,420
好了 下面是演示问题
All right, so here's the demonstration problem.
93
00:11:46,580 --> 00:11:53,630
我们有一个穿孔机 本质上是一个钢穿孔机穿透一个相对柔软的材料
We have a punch penetration, essentially a steel punch penetrating a relatively soft material.
94
00:11:53,910 --> 00:11:57,680
这包括一些可塑性 好吧
Uh, that includes some plasticity. Okay?
95
00:11:57,860 --> 00:11:59,960
我们有一个初始洞
We have, uh, an initial hole.
96
00:12:00,040 --> 00:12:03,180
呃 这是右边的变形形状
UH. This is the deform shape here on the right.
97
00:12:03,340 --> 00:12:05,160
这是视图的横截面
And this is the cross section of view.
98
00:12:05,160 --> 00:12:12,200
嗯 出于演示的目的 这实际上是一种对称分析
Um. For the purposes of this demonstration, it it's run as an actually symmetric analysis.
99
00:12:14,670 --> 00:12:16,990
好的 我把它放到这里
All right, so let me bring that up here.
100
00:12:18,590 --> 00:12:25,870
这是几何图形 一个对称模型 轴是Y轴
So here's the geometry, an actually symmetric model um, with the axis being the Y axis.
101
00:12:26,030 --> 00:12:35,700
在内部左边边缘的这个冲孔冲孔结构钢 然后你有一个更软的材料 代表板
Up the the insider left edge of this punch punches structural steel, and then you have a softer material that represents the plate.
102
00:12:37,100 --> 00:12:44,080
呃 我要做的就是打开工程数据页面 快速浏览一下
Uh. What I'll do is I'll just open up the engineering data page, will take a quick look at that.
103
00:12:44,200 --> 00:12:52,790
这种板材材料的弹性模量相对较低
Um, this plate material has a relatively low elastic modules.
104
00:12:53,020 --> 00:12:57,260
100万磅 你会指向2万磅
Uh, 1 million p. s. I. you'll point at 20000, uh p.
105
00:12:57,360 --> 00:13:01,640
s I 然后是A-A 非常柔软 屈服后 僵硬
s. I. And then A-A, very soft, uh, post yield, stiff.
106
00:13:01,830 --> 00:13:04,430
这就是所谓的切线模
Well, uh, that's called the tangent modules.
107
00:13:04,590 --> 00:13:06,450
这是10000p sI
And that's 10000 p. s. I.
108
00:13:09,130 --> 00:13:15,560
好吧 嗯 现在我想看看这个模型的几个特点
All right, HM. Now I want to look at a couple of features of this model.
109
00:13:15,900 --> 00:13:18,580
我们来看看接触界面
Uh, let's take a look at the contact interface.
110
00:13:18,740 --> 00:13:23,230
首先 冲头和平板之间有摩擦接触
1st, we have frictional contact between the punch and the plate.
111
00:13:23,630 --> 00:13:25,570
这里看不太清楚
Uh, it's a little hard to see here.
112
00:13:25,730 --> 00:13:51,630
让我们看一下 嗯 我们有这些边缘 或者说是冲孔上的目标表面 然后 嗯 我们有在抗辩上的接触面 代表 呃 这一个 和混合 嗯 在角落里
Let's look in, Um, we have these edges, or the targets surface on the punch, and then, um, we have the contact surface is over here on the plea, representing, UH, here this one, and the the blend, uh, at the corner.
113
00:13:52,890 --> 00:14:05,930
现在 有一件事需要注意 我们有一个非常高的 4es的摩擦系数 很大的偏转效应
Now, one thing to notice here is we have a very high friction coefficient of .4 es, OK, large deflection effect.
114
00:14:06,100 --> 00:14:10,160
所以如果我看分析设置 你会发现很大的情感是打开的
So if I look under analysis setting, you see large affection is turned on.
115
00:14:10,280 --> 00:14:35,050
这意味着 如果激活了几何非线性 嗯 这样他们将会占在分析 随时得到任何类型的旋转结构的 说 五度或以上 或结构的菌株 呃 超过一几 几个百分点 嗯 那么你应该总是激活大变位
What this means is that, if activated the geometric nonlinearities, um, so that they will now be accounted for during the analysis, Anytime you get any type of rotation in the structure of, say, five degrees or more, or strains in the structure of, uh, more than A-A few per-, a few percent, UM, then you should always activate large deflections.
116
00:14:35,210 --> 00:14:39,630
有时候很难判断这是否会产生效果
It it's sometimes it's difficult to tell whether or not that's going to have an effect.
117
00:14:39,750 --> 00:14:50,000
UM如果你有任何疑问 你需要打开它 因为在很多情况下 巨大的抑郁影响会对最终解决方案产生重大影响
UM If there's, if you do have any doubt, you need to turn that on, because in many cases, uh, large depression effects can have a significant effect on the final solution.
118
00:14:50,520 --> 00:14:52,840
这就是你真正需要做的
So that's all you really have to do.
119
00:14:53,220 --> 00:14:56,800
打开这个可以激活几何提名效应
Turn this on to activate geometric nominee ority effects.
120
00:14:58,640 --> 00:15:04,690
好的 嗯 我要提到的另一件事是如果我在几何中观察 2d
Okay, um. The other thing if I got to mention is if I look under geometry, the two D.
121
00:15:04,750 --> 00:15:06,630
行为设置为轴对称
behavior is set to axy symmetric.
122
00:15:08,270 --> 00:15:10,570
好了 让我们进行第一分析
All right, so let's run the 1st analysis.
123
00:15:10,730 --> 00:15:11,010
我有一个
I have a
124
00:15:11,250 --> 00:15:16,080
非常契合这个结构
very uh course mesh on the structure.
125
00:15:16,340 --> 00:15:17,790
我有固定的支柱
I have a fixed support
126
00:15:19,430 --> 00:15:26,790
在平板的外径处 我在顶部边缘施加了一个力
at the outer diameter of the plate, and I have a force applied to the top edge of the
127
00:15:27,060 --> 00:15:33,590
300磅的重量 这代表了全部的力量 呃 全部的力量在
of 300 lb. And that re- represents the the full force around, uh, the full force on the on the
128
00:15:33,810 --> 00:15:37,630
整个冲床都是360度的
entire plunge through 360 degree structure of that punch.
129
00:15:39,550 --> 00:15:43,110
好吧 这些都在一次装载中解决了
Okay? And this is all solved in a single load.
130
00:15:43,260 --> 00:15:45,780
我来解一下
So let me try to solve this.
131
00:15:49,300 --> 00:15:54,610
它运行得非常快 因为它实际上是对称分析 尽管它是非线性的
And it runs quite quickly, because it's an actually symmetric analysis, even though it's non linear.
132
00:15:57,450 --> 00:16:02,450
你知道的是 你会注意到你得到了一个解 但是位移
And what you know is, what you'll notice is that you do get a solution, but the displacement.
133
00:16:02,610 --> 00:16:05,590
所以这是完全的诽谤 是非常高的数字
So this is total defamation, is extremely high number.
134
00:16:06,500 --> 00:16:15,090
好 它告诉我那个击打没有被上下文正确地限制 它在那里
All right, so it tells me that that punch was not properly constrained by the contextor it's there Um.
135
00:16:15,250 --> 00:16:24,970
另一种检查方法是 我可以关闭弹簧周 但关闭弱弹簧并解决它
The other way we can check this is I could turn off the week Spring, but turn off the weak springs and resolve it.
136
00:16:26,370 --> 00:16:31,990
你会在消息窗口下看到一条新消息 上面写着 一个永恒的解决方案
What you'll see is you'll see a new message down in the message window that says, an eternal solution.
137
00:16:32,150 --> 00:16:42,630
震级极限是 这意味着 我们有刚体运动 刚体运动 我们需要防止这种情况发生
Magnitude Limit was a That means we have, we have rigid body motion, um, rigid body motion in the We need to prevent that from happening.
138
00:16:44,470 --> 00:16:58,090
在非线性分析中 你应该经常做的一件事是 你应该总是 尝试使用多个子步骤 这样你就可以更慢地加载
One thing you should always do in a non linear analysis, you should always, um, try to use more than one substep so you can ramp on the load more slowly.
139
00:16:59,330 --> 00:17:03,170
我要做的是 我要用不同的方式来安装它
What I'm going to do here, um, I'm actually going to install this a little differently.
140
00:17:03,460 --> 00:17:05,960
我要用两个步骤
I'm going to use two steps.
141
00:17:06,680 --> 00:17:12,070
在第一步 我要施加一个非常小的 1磅的力
In that 1st step, I'm going to apply a very small force of 1 lb.
142
00:17:12,750 --> 00:17:19,370
第二步 我要把我的300英镑全部用在潘趣酒上
The 2nd step, I'm gonna apply my full £300 for to the punch.
143
00:17:20,170 --> 00:17:24,150
这里我想这1LB
In this I'm thinking that that 1 LB.
144
00:17:24,310 --> 00:17:27,610
力可能会把它推到接触面上
Force might be able to push it into contact.
145
00:17:27,810 --> 00:17:29,350
我们看看会发生什么
So let's see what happens there.
146
00:17:29,370 --> 00:17:35,830
现在我把它作为两步分析 两步或一磅
So now I'm running it as a two step analysis, two steps or step 1 lb.
147
00:17:35,990 --> 00:17:38,210
我可以走300磅 我们来解一下
So I could step 300 lb. So let's solve that.
148
00:17:42,750 --> 00:17:47,690
再一次 我们得到了内部溶液的大小极限
And again, we get an internal solution magnitude limit exceeded.
149
00:17:47,850 --> 00:17:50,630
这又告诉我们有刚体运动
So that again tells me that we have rigid body motion.
150
00:17:51,470 --> 00:18:03,990
好了 你会注意到我们在开始的时候有一个间隙 两个部分之间有一个间隙 在冲孔和盘子之间
All right, um, what you'll notice here is we have a gap between starting, gap between the two parts, between the punch and the plate.
151
00:18:04,910 --> 00:18:12,540
所以我检查这个的方法是 如果我有一个复杂的模型 可能很难找到这些缺口
So the way I can check for this, if I have a complicated model, it might be difficult to find these gaps.
152
00:18:12,710 --> 00:18:14,610
这样我就能更容易找到它们
So I can find these more easily.
153
00:18:14,630 --> 00:18:24,570
如果我打开UH连接 插入联系工具 我就能得到每个联系对的初始信息
If I go to go to UH connections, insert contact tool, and I can get some initial information about each of my contact pairs.
154
00:18:24,730 --> 00:18:27,070
我生成了初始接触结果
So I generate initial contact results.
155
00:18:29,870 --> 00:18:39,460
它必须通过一些计算来检查初始环境状态如果我看到除了灰色或白色以外的任何东西
And it has to run through some calculations to check for the initial context status if I see anything other than grey, uh or white.
156
00:18:39,620 --> 00:18:42,800
这意味着我们有一个开放接触条件
What it means is that we have an open contact condition.
157
00:18:43,180 --> 00:18:46,760
如果是一个开触点条件 加上那个触点对 这就是间隙
If an open contact condition, add that contact pair, and this is the gap.
158
00:18:47,140 --> 00:18:52,740
所以有两个选择 我可以承受这一拳然后移动这么多来缩小差距
So have two options. I can either take that punch and move it by this amount to close that gap.
159
00:18:52,900 --> 00:18:57,400
如果我可以回到设计模型或者我的CAT课程 我就可以离开
If I could go back to design model or my CAT program, I could move apart.
160
00:18:57,480 --> 00:19:06,910
我还可以在联系人设置下使用一个技巧
Or another thing I can do here, use a trick under my contact settings.
161
00:19:07,790 --> 00:19:16,320
我可以去到界面处理的几何修改 我可以改变这个来调整触摸
I can go down to the interface treatment under Geometric modification, and I can change this to adjust a touch.
162
00:19:16,480 --> 00:19:28,230
现在 有效地 这将会做什么它将会取我的接触面并用最小间隙边来抵消它 以缩小两个部分之间的间隙
Now, effectively, what this is gonna do it's gonna take my contact surface and offset it by the minimum gap side to close that gap between the two parts.
163
00:19:30,550 --> 00:19:38,410
我应该做的另一件事 你应该总是在非线性分析中这样做 你应该使用不止一个子步骤
The other thing I should do, you should always do this in a in a non linear analysis, you should use more than one substep.
164
00:19:39,400 --> 00:19:46,080
现在 对于我的第一步 我不需要这么做 因为我有一个非常小的力供给
Right now, for my 1st step, I don't really have to do it, because I have a very small force supply.
165
00:19:46,450 --> 00:19:50,790
现在 我的第二步 我要再增重299磅
Now, my 2nd step, I'm applying an additional 299 lb.
166
00:19:51,050 --> 00:19:57,660
所以我想用多个子步骤 它们会它们会对力倾斜 而不是施加所有的力
So I'd like to use multiple substeps, which will which will then ramp on the force, instead of applying all that force.
167
00:19:57,820 --> 00:20:03,570
其中一步 要用多个子步来增加力
And one step, uh, it's going to use multiple subteps to ramp on that force.
168
00:20:04,410 --> 00:20:09,670
为了做到这一点 我进入分析设置 把步骤数改为2
So to do that, I go to analysis settings, change my step number to two.
169
00:20:10,310 --> 00:20:13,030
我要打开自动计时
I going to turn on automatic time
170
00:20:13,330 --> 00:20:17,550
步 然后有子步
stepping, and then I have initial subteps,
171
00:20:17,940 --> 00:20:20,020
最小次步和最大
min subteps and max.
172
00:20:20,460 --> 00:20:22,830
这有什么用
What this is gonna do
173
00:20:23,380 --> 00:20:30,510
我们用最初的20步来增加 300磅的力
is we use initially 20 sub steps to ramp on that, uh, that force of 300 lb.
174
00:20:30,920 --> 00:20:35,270
如果它迅速收敛
If it converges rapidly, um,
175
00:20:35,460 --> 00:20:41,100
它可以使用更少的子步骤 实际上我可以把这个数写小一点 比如10
it could use fewer substeps. So I could actually make this a smaller number, let's say ten.
176
00:20:41,260 --> 00:20:46,740
然后 如果它转换得很快 就会放大后面每一个子步骤的大小
And then, if it converted rapidly, will then enlarge the size of each subsequent substep.
177
00:20:46,980 --> 00:20:57,330
直到它达到最小数量的十个子集 如果它难以收敛 自动时间步进器就会使用更多的物质
And and until it reaches a minimum number of subsets of ten, um, if it has difficulty converging, the automatic time stepper will then use more substance.
178
00:20:57,490 --> 00:20:59,490
所以最好使用一系列的子seps
So it's good to use a range of subseps.
179
00:20:59,650 --> 00:21:04,100
这里的初始子集是每个子集的初始大小
Here the initial subsets is the starting size of each subset.
180
00:21:04,260 --> 00:21:06,240
所以它要带走300磅
So it's going to take my 300 lb.
181
00:21:06,510 --> 00:21:09,310
再细分为20种物质
Subdivide that load into 20 substance.
182
00:21:10,030 --> 00:21:18,800
现在我也有了这个选项 我改变了触摸来缩小 缩小最初的差距
And I also now have this option that I changed just to touch to close that, close that initial gap.
183
00:21:19,480 --> 00:21:23,150
然后我就可以生成初始接触结果
I can then generate initial contact results.
184
00:21:24,710 --> 00:21:35,380
现在 如果我看一下结果 我有一排白色的信息
And now, if I look at the results there, I have a nice, uh, white row of information.
185
00:21:35,900 --> 00:21:39,770
接触元素的数量是1
Uh, number of contacting elements is one.
186
00:21:40,250 --> 00:21:43,790
最初的穿透量很小
The um, initial penetration is a tiny number.
187
00:21:43,950 --> 00:21:50,910
不再有空隙了 所以最初的 触碰这个选项 有效地填补了这个空白
There's no gap anymore. Uh, so the initial, uh, the just a touch option, effectively closed that gap.
188
00:21:51,070 --> 00:21:52,690
现在让我试着运行它
So now let me try and run it.
189
00:21:56,450 --> 00:22:00,170
当它运行时 我可以观察力的收敛行为
While it's running, I could look at the force convergence behavior.
190
00:22:00,350 --> 00:22:04,410
它在溶液输出下 我可以把它设为第四收敛
It's under solution output. I can ch- set this to 4th convergence.
191
00:22:05,170 --> 00:22:15,160
这条紫色的线是作用力和内力的差值
This purple line is the uh difference between the applied force and the internal force.
192
00:22:15,160 --> 00:22:22,160
一旦紫色的线低于浅蓝色的线 浅蓝色的线就是收敛容忍数
It is that once this purple line gets below the light blue line, the light blue line is the convergence tolerance number.
193
00:22:22,280 --> 00:22:27,820
它需要低于这个数才能有收敛子步长
It needs to get below that number to have a converge substep.
194
00:22:27,980 --> 00:22:33,180
每一条绿虚线 都代表一个收敛子步骤
Each of these dash green lines, uh, represents a converge substep.
195
00:22:34,220 --> 00:22:45,420
看下面 它们都收敛到第2次loadstep的80%以上
Okay, so now, if I look down at the bottom, they're both converge up to just over 80% of that 2nd loadstep.
196
00:22:46,700 --> 00:22:49,260
现在 发生了什么事 让我们一起来看看吧
Now, what happened? Let's take a look.
197
00:22:49,620 --> 00:22:55,970
如果我往下看信息下面 上面写着 托盘引擎无法汇合
If I look down under the air uh, under the messages, it says, the salver engine was unable to converge.
198
00:22:56,410 --> 00:22:58,430
呃 所以这不是很有用 对吧
Uh. So that's not very helpful, right?
199
00:22:58,590 --> 00:23:00,450
它没有给我们很多信息
It doesn't give us a lot of information.
200
00:23:01,010 --> 00:23:12,460
获得更多的信息 我可以从力 力的收敛图 转换到解算器的输出 一直到最下面 看看这些信息
Get more information. I can switch from force, the force convergence graph, to solver output, going down to the bottom, and look at the messages.
201
00:23:13,100 --> 00:23:19,980
下面这个消息说 元素02 45被高度扭曲了 所以这是一个元素扭曲错误
This message down here says, element 02:45 has become highly distorted, so this is a element distortion error.
202
00:23:20,700 --> 00:23:28,740
好的 我可以看看结果 这是对于非收敛解看起来很糟糕 对吧
All right, I can look at the results, and this is for the unconveged solution that looks horrible, right?
203
00:23:28,910 --> 00:23:42,510
严重严重的无序元素 我可以看一下在那之前的元素解 所以我可以选择最后一个收敛点 得到那个看起来不错的结果
Severely severely disordered element. I could look at the elements solution prior to that, so I could pick that last converged point, retrieve that result that actually looks fine.
204
00:23:45,830 --> 00:23:53,450
好吧 让我看看 我们将如何克服这一点
Okay, um, let's see here. How are we going to overcome this?
205
00:23:53,760 --> 00:24:05,210
我们可以做的一件事是如果我观察初始网格 角落的这个元素是其中的一个元素 对吧
Well, one thing we can do if I look at the initial mesh, this element at the corner is a single element in there, right?
206
00:24:05,230 --> 00:24:09,790
混合本身有两个方面
And the the blend itself has two sides to it.
207
00:24:10,090 --> 00:24:12,590
这在语境区不是一个很好的元素
It's not a very good element in the context zone.
208
00:24:13,430 --> 00:24:18,630
嗯 我要做的就是改变网格
Um? So what I'm going to do is I'm going to change that mesh.
209
00:24:21,230 --> 00:24:25,020
我要激活或释放这个练习
I'm going to activate or unsuppressed this exercising.
210
00:24:25,180 --> 00:24:35,530
这个练习指定了数字 指定了五个元素混合或重新生成网格
This exercising specifies the number, specifies, uh, five elements around that blend or regenerate the mesh there.
211
00:24:37,330 --> 00:24:38,750
现在我要解
And now I'm going to resolve.
212
00:24:38,910 --> 00:24:41,310
在我决定之前 让我先做一件事
Now, before I resolved, let me just do one thing here.
213
00:24:41,470 --> 00:24:43,210
你看这个牛顿说唱和残余
You see this Newton raps and residual.
214
00:24:43,790 --> 00:24:48,230
我要在这里放一个数字 这里是1
I'm gonna I'm going to put a number in here, in this case, one.
215
00:24:48,650 --> 00:25:01,490
这样我就可以画出 网格中力和平衡的边缘图 或者等高线图 如果不收敛 这就很重要了
And what that's going to allow me to do is going to allow me to plot, create a fringe plot, or contour plot of the force and balance in the mesh uh, and that's going to be important if it doesn't converge.
216
00:25:04,190 --> 00:25:14,710
我们再解一次 我可以考虑收敛
Let's solve this again, and I could look at the or with convergence
217
00:25:14,940 --> 00:25:21,220
的行为 当它在这里运行的时候 它在向前移动
behavior. While it's running here, it's moving along.
218
00:25:21,240 --> 00:25:27,250
我需要这个时间点达到2的值 因为第2次加载 默认步骤 有
I need this time point to get up to a value two, because the 2nd load, step by default, has
219
00:25:27,370 --> 00:25:29,470
时间价值也与之相关
a time value too associated with it.
220
00:25:32,670 --> 00:25:36,150
好吧 它达到了
Okay? And it gets up to
221
00:25:37,710 --> 00:25:39,350
聚合点 所以
the converged point. So
222
00:25:39,550 --> 00:25:45,510
太好了 嗯还有 所以
that's great, um and all. So
223
00:25:45,710 --> 00:25:48,910
我所要做的就是修改网格
all I had to do there was modify the the mesh.
224
00:25:49,070 --> 00:25:52,890
在这种情况下 我可以看看总的变形量
In the context, I can look at the total Deformation.
225
00:25:54,210 --> 00:25:57,290
让我看看最后一个时间点
Let me look at the final time point here.
226
00:25:58,680 --> 00:26:04,700
这是总变形和等效塑性应变
That's the total Deformation and the equivalent plastic strains.
227
00:26:04,720 --> 00:26:07,760
这里的塑性应变水平很高
We have a very high plastic strain level in here.
228
00:26:10,090 --> 00:26:17,320
好吧 我们做点别的吧
Okay, um, let's do something else.
229
00:26:17,480 --> 00:26:22,340
是的 我想确认一下 嗯
Yeah, I want to check. Um.
230
00:26:24,740 --> 00:26:33,200
好吧 现在 很好 现在让我看看我们能不能 我们能把这个潘趣酒往下推
Okay. Now that's fine. So let me now see if we can, we can push this punch further down.
231
00:26:33,360 --> 00:26:37,690
所以我要把第二步变成400磅
So I'm going to change that 2nd step to 400 lb.
232
00:26:39,810 --> 00:26:41,930
我会试着解决这个问题
And I'm gonna try and resolve it.
233
00:26:52,050 --> 00:27:02,110
现在 这些红色的线在c-4融合历史图中 这些代表的 呃 是双性的
Now, these red lines here in the in the c- 4th Convergence History plot, what these represents, uh, represent, are bisexued.
234
00:27:02,390 --> 00:27:12,590
这就意味着 在这一个子步骤中 收敛起来非常困难所以它将子步骤分成了两半
This means that at this substep, it had so much difficulty converging that it actually subdivided that substep in half.
235
00:27:13,310 --> 00:27:20,900
到那个时候 它就能进一步增加负荷
And, uh, at that point it, UM, it's able to ramp up the load further.
236
00:27:22,060 --> 00:27:26,020
好的 嗯 自动时间步进器会处理所有这些
Okay, uh. And the automatic time stepper takes care of all that.
237
00:27:26,890 --> 00:27:31,090
好了 再一次 我们有一个不收敛的问题
All right, so once again, we have a non convergence problem.
238
00:27:31,610 --> 00:27:35,820
嗯 我们来看看一些事情
Um? And so we take a look at a couple of things.
239
00:27:35,820 --> 00:27:38,740
我可以看看牛顿的说唱和残余力
I can look at the Newton raps and residual force.
240
00:27:39,020 --> 00:27:44,820
这里有一个牛顿撞击和res-在机会子的另一边有非常高的剩余力
And there's a Newton raps and res- very high residual force right there on the other side of the chancer.
241
00:27:46,900 --> 00:27:52,100
嗯 我可以做的另一件事是观察解的输出
Um. The other thing I can do is I can look at the solution output.
242
00:27:52,260 --> 00:28:02,320
所以我把力收敛转换到解算器的输出 然后一直到底部 我们又得到了一个元素畸变误差
So I switch force convergence to solver output, and go down to the bottom, and again, we have an element distortion error.
243
00:28:02,580 --> 00:28:04,760
只有02 25被高度扭曲了
Only 02:25 has become highly distorted.
244
00:28:06,240 --> 00:28:09,080
好吧 那我们该怎么办
All right, so what are we going to do about this?
245
00:28:09,240 --> 00:28:19,580
嗯 我注意到一件事当我当我设置我的接触器时 我没有把接触器的边缘延伸到盘子的顶部
Well, one thing I noticed when I when I set up my contacts, I didn't extend this contact edge around the top side of the plate.
246
00:28:19,940 --> 00:28:29,110
所以我应该做的是改变接触面让它延伸到平板的顶部 所以我添加了额外的边
So what I should do is I should change my contact surface that it extends over to the top of the plate, so I add that additional edge.
247
00:28:30,270 --> 00:28:35,380
还有一件事对接触分析有帮助
Um Another thing that will help in a contact analysis.
248
00:28:36,260 --> 00:28:41,190
嗯特别 当你有软化行为时 这尤其是个问题
UM Special. This is especially an issue when you have softening behavior.
249
00:28:41,350 --> 00:28:50,250
所以我们有很多的可塑性在这个位置发生 所以材料软化了 嗯 很多
So we have a lot of plasticity that's occurring at this at this uh, chance for location, so the material softening, uh, quite a bit.
250
00:28:50,810 --> 00:28:55,880
为了帮助它收敛 我们可以改变这个法向刚度值
To help this to converge, we can change this normal stiffness value.
251
00:28:55,880 --> 00:28:58,760
我可以把它改成程控 改成手动
I can change it for program control, to manual.
252
00:29:00,420 --> 00:29:07,720
不好意思 让我想想 曼纽尔 正常的刚度系数 由于错误 会被设为1
Oh, I'm sorry, uh, let's see, manuel, the normal stiffness factor, by the fault, will be set to one.
253
00:29:08,100 --> 00:29:13,960
把这个改成 1 这会软化那个位置的语境
We change this to .1. That's going to soften the context ifness at that location.
254
00:29:14,290 --> 00:29:25,680
再次强调 这一点很重要当你在柔性体之间进行接触或者在有塑性或其他材料软化行为的地方进行接触时
Again, that's going to be important when you have contact between flexible bodies or contact um in a location where you have, uh, plasticity or some other material softening behavior.
255
00:29:28,240 --> 00:29:40,570
好 现在重新运行它 我们会再看看力的收敛历史
All right now and rerun it, and we'll look at the force convergence history again.
256
00:29:44,540 --> 00:29:59,150
是的 看起来很像 一开始 大概是72% 79%
Yeah, looks pretty similar. In the beginning, it's about 72 %, 79 %.
257
00:29:59,310 --> 00:30:05,110
在第二步中 92%的人都很粗糙
Of that 2nd step ruggles 92 %.
258
00:30:07,870 --> 00:30:10,510
但它并没有成功
And it doesn't quite make it.
259
00:30:10,950 --> 00:30:16,190
好吧 再一次 我们可以看看求解器的输出
Okay, again. So we could look at the uh Solver output.
260
00:30:20,110 --> 00:30:22,090
一次 我们有一个扭曲的元素
Again. We have a distorted element.
261
00:30:22,250 --> 00:30:23,990
看起来是相同的元素
Looks like the exact same element.
262
00:30:24,230 --> 00:30:25,090
我可以看看
I can look at the
263
00:30:25,370 --> 00:30:28,490
残余势力 看起来非常非常相似
residual force, it looks pretty, pretty similar.
264
00:30:28,550 --> 00:30:39,200
现在 这告诉你的是 我不担心力的价值 但它它给了你一个位置的概念 导致了 比如 处女问题
Now, what this is really telling you, uh, I wouldn't worry about the value of the force, but it it gives you an idea of the location at that's causing, like, a virgin problem.
265
00:30:40,760 --> 00:30:42,870
所以这个这个
So this this, uh,
266
00:30:43,180 --> 00:30:45,670
接触对的这一部分 呃 这一部分
this part of the contact pair, UH, this
267
00:30:45,990 --> 00:30:48,450
接触区的区域造成了收敛问题
region of the contact zone is causing that convergence problem.
268
00:30:48,490 --> 00:30:51,870
嗯 我们可以做的一件事是
Um. So one of the things we can do here,
269
00:30:53,310 --> 00:31:07,960
如果我们能改变接触方式 配方 接触算法 默认情况下 它使用增强液体法 这通常是你想要开始的地方
UH, if we can change the contact, UH formulation, the contact algorithm, the By default, it uses an augmented liquorage method, which is generally the the place that that you want to start.
270
00:31:08,120 --> 00:31:13,440
它通常是最健壮的 最好的上下文和准确性的结合
It's generally the most robust and the best combination of context in this and accuracy.
271
00:31:13,600 --> 00:31:17,280
看 这里还有其他几个选择
See, there are a couple other options here.
272
00:31:17,440 --> 00:31:21,000
其中一个值得尝试的是普通的拉格朗日算法
One of them that's worth trying is the normal lagrange algorithm.
273
00:31:21,160 --> 00:31:30,530
这可以防止任何接触在接口之间的渗透 并利用接触压力来加强接触兼容性
This prevents any penetration, uh between the interfaces, and uses the contact pressure to enforce contact compatibility.
274
00:31:31,810 --> 00:31:35,730
这当然是我们在这类问题中尝试的
So this is certainly we're trying in these types of problems.
275
00:31:36,100 --> 00:31:50,630
嗯 现在我要解决它 我要把这个切换回强制转换 我们看看会发生什么
Um? So now I'll resolve it, and I'll switch this back over to force conversion, and we'll see what happens.
276
00:31:51,080 --> 00:31:58,360
我们得到50% 60% 73%
We get to 50 %, 60 %, 73 %.
277
00:31:58,520 --> 00:32:01,560
你可以看到 都在努力趋同
You can see is struggling to converge.
278
00:32:01,720 --> 00:32:08,980
我们有很多 平衡缓和 最后 它最终会收敛
We have a lot of, uh, equilibrium moderations, and then finally, it does eventually converge.
279
00:32:09,790 --> 00:32:16,450
这就是收敛解 现在 你会注意到一件事 这很有趣 我们确实有一个似乎是永久性的空白
This is our converged solution. Now, one thing you'll notice, which is kind of interesting, is we do have what appears to be a permanent gap in there.
280
00:32:16,470 --> 00:32:28,440
那是因为我们改变了这个接触的方法并调整为接触 所以它用间隙值抵消了接触界面
And that's because we changed this contact uh method to and adjust to touch, so it offset that contact interface by the gap value.
281
00:32:28,500 --> 00:32:33,940
所以你总能看到这两个界面之间有一个几何间隙
So you will always see that what appears to be a geometric gap between those two interfaces.
282
00:32:34,100 --> 00:32:42,900
如果你不想这样 你就必须手动移动 接触中最初等效塑性应变的两个部分
If you don't want that, you're going to have to manually move the, uh, the two parts in the contact initially equivalent plastic strain.
283
00:32:44,060 --> 00:32:49,470
我们做了很多事情来得到收敛解
Okay, so there were a number of things we did to obtain and converge solution there um.
284
00:32:49,630 --> 00:33:07,880
如果我应该提一下 如果我们想要p-把这个 啊 这个穿孔器推得更远 所以它到达这个外缘 这个穿孔器的外缘 你实际上要把这个从力载荷变成位移载荷
If one that I should mention, if we want to p- push that, ah, that punch further through, so that it reaches the this outer edge, this outer edge of the punch, you would actually have to change this from a force load to a displacement load.
285
00:33:08,080 --> 00:33:33,580
这样做的原因是 如果你继续施加额外的力 最终会发生的是整个部分会穿过 整个冲力会穿过 没有任何东西可以约束它 一旦它到达一旦整个部分到达这个冲孔的外径
Um, The reason for that is because if you continue to apply an additional, additional forces to this, eventually what's gonna happen is the the entire section will push through, the entire punch will push through, and there will be nothing to restrain it, um at um once it once it reaches once the entire section reaches the outer diameter of this punch.
286
00:33:33,740 --> 00:33:38,580
所以你必须转换成位移强迫载荷
So you would have to switch over to a displacement enforced load.
287
00:33:39,820 --> 00:33:45,080
好的 我们用了更多的子步骤来帮助我们得到收敛性
Okay, so we used, um, more substeps to help us obtain convergence.
288
00:33:45,240 --> 00:33:48,080
我们关闭接触间隙使用调整触摸选项
We close the contact gap using the adjusted touch option.
289
00:33:48,240 --> 00:33:55,960
我们改进了接触区域的网格我们减少了环境刚度
We refine the mesh in the contact zone, um We reduce context stiffness.
290
00:33:56,120 --> 00:34:10,640
当我们有软化行为时 我们修改了接触界面 将平板顶部的边缘加入到接触对中 我们切换到正常的车库配方
That helps UH When we have softening behavior, we modified the contact interface by adding that edge at the top of the plate to the contact pair, and we switched over to a normal garage formulation.
291
00:34:13,320 --> 00:34:17,880
好了 让我们回到刚才的演讲 我想在这里结束
All right, so let me go back to the presentation I just want to wrap up here.
292
00:34:17,920 --> 00:34:24,700
嗯 因此 收敛问题 它们通常可以用以下方法克服
Um. So, convergence problems, they can often be overcome using the following methods.
293
00:34:24,860 --> 00:34:28,240
我们可以添加子步骤 那是那是你应该经常尝试的事情
We can add subteps. That's that's something you should always try.
294
00:34:28,260 --> 00:34:34,060
嗯 打开附加子步骤的自动计时来慢慢增加负荷
Uh see, turn on the automatic times that are at additional substeps to slowly ramp on the loads.
295
00:34:34,700 --> 00:34:38,700
我们可以确保没有初始间隙接触界面
We can ensure that there are no initial gaps contact interfaces.
296
00:34:38,860 --> 00:34:40,960
这有助于防止刚体运动
That helps to prevent rigid body motion.
297
00:34:41,360 --> 00:34:44,580
呃 我们可以去除尖角和接触区域
Uh? We can remove sharp corners and contact regions.
298
00:34:44,740 --> 00:34:47,160
通过加入混合物或菲利普斯
Um by adding blends or phillips.
299
00:34:47,250 --> 00:34:53,050
我们可以在经历软化或结构灵活性的区域降低上下文刚度
We can reduce context stiffness in regions that experience softening or structural flexibility.
300
00:34:54,020 --> 00:35:02,530
呃 我们可以调整或细化网格 以尽量减少元件变形 并改善我们想要的接触压力分布
Uh? We can adjust or refine the mesh to minimize element distortions and improve contact pressure distribution we want to have.
301
00:35:02,910 --> 00:35:07,130
最好有更多的元素相互接触 而不是更少的元素
It's better to have more elements that are in contact with each other, rather than fewer elements.
302
00:35:07,490 --> 00:35:10,070
我们还可以尝试另一种接触算法
And we can also try another contact algorithm.
303
00:35:10,230 --> 00:35:19,070
呃 除了标准的增强lagron 你可以切换到普通的车库 或者还有另一个选择 你可以尝试基于投影的联系 这通常也很有帮助
Uh. Instead of the standard augmented lagron, you can switch over to normal garage, or there's another option, which is projection based contact that you can try, and that's often helpful as well.
155-2731-8020
