【讨论】请问关于三角形壳单元出现的问题？

wangscore

用Patran和Nastran做了一道静力学的例题，单元为三角形壳单元，共计4个单元，6个节点，单元排序相同。只是节点排序不同，一种是
（1）1，2，3  
（2）5，3，2
（3）2，6，5
（4）3，5，4
另一种是
（1）2，3，1
（2）3，2，5
（3）6，5，2
（4）5，4，3
计算结果位移值相同，而应力值却不同，请问各位为什么会这样？f06结果文件中表示的应力值是什么位置的应力值？为什么没有节点的应力值，而在四边形单元时却有？

wangscore

MSC.Patran Discussion Group
Mr. Wang
You changed element local coordinates when you changed nodal connection
sequence, therefore principal stresses were changed. But if you post results
with von mises stress, you will see no difference.
All nodal stresses were extrapolated from element stresses. check
NASTRAN USER MANUAL, you will find what you want!
  
Guijun Deng
CAE Technical Specialist
Superior industries, International
  
-----Original Message-----
From: wangsg@ciomp.ac.cn [mailto:wangsg@ciomp.ac.cn]  
Sent: Sunday, April 11, 2004 7:48 PM
To: MSC.Patran Discussion Group
Subject: [msc.patran] about tria3 shell problem
  
MSC.Patran Discussion Group
Hi,Everyone:
there are two structure statics bdf files in the inclosure,both of
elements are tria3 shell, only difference is node sequence:
(1）1，2，3
（2）5，3，2
（3）2，6，5
（4）3，5，4
another:  
（1）2，3，1
（2）3，2，5
（3）6，5，2
（4）5，4，3
we can get same displacement by Nastran,howerver, stress results are
different.Can anyone tell me what is the reason of it? what is the meaning
of stress result in the f06 file,why not node stress values?
in postprocessing stress tensor graph has a difference with .rpt file?
thanks.
Best Regards