急!请教一个关于重启动 材料状态设置的问题.在线等.

citybeggar

请问诸大侠,
在重启动分析中,
如何维持上一步分析中的位移和应变,
而叫结构中的应力都为0.
即分析之初,位移和应变均不为0,
而应力全部为0.
能实现么?如何设置关键字?

或者说,
在某一个状态,好象是不是重启动分析也不重要,
如何使abaqus分析中的应力状态全部清0??
有这样的关键字么??


又即,不写UMAT能实现么?
如果非要写UMAT,有什么建议么?
哪位大侠不吝赐教,不胜感激.
先谢谢了..

yogayoga

There are solutions for different situation. Need to know a little bit more about what you want to do.

citybeggar

yogayoga wrote:
There are solutions for different situation. Need to know a little bit more about what you want to do.

在上一个显式分析中,
我得到了一些节点位移应力应变状态,
在下一个隐式分析中,
我想恢复节点位移应变状态,
但要求应力都为0.......

有什么好的法子么??

citybeggar

卓越太阳 wrote:
唉，我和你恰恰相反，我要求应力不变，位移为零

   
    在ansys中,很容易实现,应力不变,位移为0
有一个upcoord命令..
    如果你会ansys的话,不妨试一下..

   在abaqus里面,有一个思路,不妨试一下,
*IMPORT, UPDATE=YES,这时,分析的初态是当前态,
就是位移和应力都为0的意思,
这个时候,你固定住所有的节点的自由度,
然后在每一个节点上施加上一个分析中获得的节点力,可编程序提出来,
不过,这个时候你一定施加那些节点力的反力,就是以前是10牛,现在
应该-10牛,这样,计算出来的应力状态,才是你要的,
然后下一个分析中,
释放所有的节点的自由度,当然得保留一些你要的约束,
这时应该就是你要的状态了..

   不过,应力怎么样设为0呢??
怎么样,消除残余应力呢?而又保持位移位移应变状态不变呢??
积存在单元中的能量怎么释放出来啊??

citybeggar

pumahiker wrote:
晕，应变不清零，应力清零？
不知道你的物理过程是什么。

大哥说的好有道理啊,
^$^%((^&%^$,可一个外国的牛鼻子论文里面就是这么说的,
头疼,
不知道他怎么搞出来的,
我想重现在它的过程,然后再深化一下,看有没有得搞头.....

大家帮我想一下..

yogayoga

In this case, there is a simple solution. You can check *anneal. It will get rid
of all the stress and internal variable, field variables and keep strain and displacement.

For the situation of keeping stress without strain, if you give a little bit more
background, perhaps there is also simple solution.

citybeggar

yogayoga wrote:
In this case, there is a simple solution. You can check *anneal. It will get rid
of all the stress and internal variable, field variables and keep strain and displacement.

For the situation of keeping stress without strain, if you give a little bit more
background, perhaps there is also simple solution.

Thank you very very much!
看来还是自己的帮助看的不到家啊,回去继续努力....
应力不不变,位移为0,
有哪些关键字可以处理呢?

我先回去试验下我的,看成不成,
回来再请教..........

真的很感谢....

yogayoga

If you think about it, keep stress without strain should not be a problem.
If stress is in equilibrium state, with proper mesh (initial mesh as deformed
coordinates) and proper boundary condition, map solution should easily
achieve that.  

If stress is not in equilibrium state already, it is physically meaningless to
achieve such a state without strain.

By the way, I'm not a civil engineer. Don't know if there are any
significances for such problem.  Anyone likes to explain the physics behind it?

citybeggar

yogayoga wrote:
If you think about it, keep stress without strain should not be a problem.
If stress is in equilibrium state, with proper mesh (initial mesh as deformed
coordinates) and proper boundary condition, map solution should easily
achieve that.  

If stress is not in equilibrium state already, it is physically meaningless to
achieve such a state without strain.

By the way, I'm not a civil engineer. Don't know if there are any
significances for such problem.  Anyone likes to explain the physics behind it?

我也不是搞土木的,
我想,
物理上应该不会出现这样一种畸形的状态,
而之所以,
纯粹是为了对某些成形过程进行优化
以期能得到一些常规的方法得不到的东西....

yogayoga兄,真是高人,
有机会还要请你多多指点才是,
真的真的非常感谢你的帮助....