声振论坛

 找回密码
 我要加入

QQ登录

只需一步,快速开始

查看: 1559|回复: 8

[求助]ansys版本不同在计算上的影响??

[复制链接]
发表于 2006-6-20 13:07 | 显示全部楼层 |阅读模式

马上注册,结交更多好友,享用更多功能,让你轻松玩转社区。

您需要 登录 才可以下载或查看,没有账号?我要加入

x
我最近用8.0 和9.0 版本算了一个模型,用的是同一组命令流,结果第一主应力一个是2兆帕多,一个是3兆帕多。 后来觉得奇怪,又换了一个模型,还是用的同一组命令流,结果两个版本算出的还是相差很多。<BR>请问各位高手,这应该是什么原因造成的???<BR>(线性,静力,实体,同一种材料,网格是自己划分)
回复
分享到:

使用道具 举报

发表于 2006-6-20 14:50 | 显示全部楼层
共同关注ing
发表于 2006-6-21 07:05 | 显示全部楼层

回复:(daviddaiya)[求助]ansys版本不同在计算上的影...

应该不会存在这样的问题<BR><BR>建议把你的例子贴出来大家分析一下
 楼主| 发表于 2006-6-21 15:27 | 显示全部楼层

谢谢

<P>好的。网格全部为扫略划分,荷载是对转向管上表面30度内加表面力。命令流如下:<BR>/clear<BR>/prep7<BR>/title,the analysis of deviator with reinforcment <BR>et,1,solid45<BR>mp,dens,1,2600<BR>mp,ex,1,3.5e10<BR>mp,prxy,1,0.1667<BR>k,100,-0.37,-2.089863,-3.2<BR>k,101,-0.37,-2.1313,-3.5<BR>k,102,-0.37,-2.1505,-4.0<BR>larc,100,102,101</P>
<P>kwpave,100<BR>csys,4<BR>circle,100,0.06,,,360,24<BR>csys,0<BR>lcome,7,8<BR>allsel<BR>lsel,s,,,2,6,1<BR>lcome,all<BR>allsel<BR>lsel,s,,,9,13,1<BR>lcome,all<BR>allsel<BR>lsel,s,,,14,19,1<BR>lcome,all<BR>allsel<BR>lsel,s,,,20,25,1<BR>lcome,all<BR>allsel<BR>lsel,s,,,2,20,1<BR>cm,huani,line<BR>cmsel,s,huani<BR>lsel,s,,,huani<BR>al,huani<BR>allsel<BR>vdrag,1,,,,,,1</P>
<P>!产生关键点<BR>k,1001,0,0<BR>k,1002,-0.09,0<BR>k,1003,-0.47,0<BR>k,1004,-0.8,0<BR>k,1005,-1.1,0<BR>k,1006,-1.1,-0.15<BR>k,1007,-0.8,-0.15<BR>k,1008,-0.47,-0.15<BR>k,1009,-0.09,-0.23<BR>k,1010,-0.09,-1.86<BR>k,1011,-0.27,-2.02<BR>k,1012,-0.27,-2.30<BR>k,1013,-0.09,-2.30<BR>k,1014,0,-2.30<BR>k,2015,-0.09,-1.50,-3.3<BR>k,2016,-1.1,-1.50,-3.3<BR>k,2017,-1.1,-1.80,-3.2<BR>k,2018,-0.09,-1.80,-3.2<BR>k,2019,-1.1,-2.30,-3.2</P>
<P><BR>!进行关键点的对称复制(按抛物线复制以便生成体)</P>
<P>kgen,2,1001,1009,1,0,0,-3.3,1000<BR>kgen,2,1010,1012,1,0,0,-3.2,1000<BR>kgen,2,2015,2016,1,0,0,-0.4,1000<BR>kgen,2,2017,2019,1,0,0,-0.6,1000<BR>kgen,2,1000,1014,1,0,0,-7,2000<BR>l,2008,2009<BR>l,2008,2007<BR>l,2007,2006<BR>l,2006,2016<BR>l,2016,2015<BR>l,2015,2009</P>
<P>l,2018,2017<BR>l,2017,2019<BR>l,2019,2012<BR>l,2012,2011<BR>l,2011,2010<BR>l,2010,2018<BR>al,16,17,18,19,21,22<BR>al,23,24,25,26,27,28<BR>vext,8,,,,,-0.4<BR>vext,9,,,,,-0.6<BR>vsbv,3,1<BR>v,1001,1002,1013,1014,3001,3002,3013,3014<BR>v,1002,1003,1008,1009,3002,3003,3008,3009<BR>v,1003,1004,1007,1008,3003,3004,3007,3008<BR>v,1004,1005,1006,1007,3004,3005,3006,3007<BR>v,1010,1011,1012,1013,3010,3011,3012,3013<BR>v,2015,2016,2017,2018,3015,3016,3017,3018</P>
<P>allsel<BR>vglu,all<BR>nummrg,all !整理编号 <BR>numcmp,all</P>
<P>!选泽箱体并划分网格<BR>allsel<BR>vsel,s,,,8<BR>vplo<BR>esize,0.1<BR>vsweep,all<BR>allsel<BR>vsel,s,,,7<BR>vplo<BR>esize,0.05<BR>vsweep,all<BR>!选择转向快中部单元划分<BR>allsel<BR>vsel,s,,,2<BR>vplo<BR>esize,0.03<BR>vsweep,all,25,26<BR>!选择转向快上下部分进行单元划分<BR>allsel<BR>vsel,s,,,3<BR>vplo<BR>esize,0.05<BR>vsweep,all<BR>allsel<BR>vsel,s,,,1<BR>vplo<BR>esize,0.05<BR>vsweep,all<BR>!选泽箱体并划分网格<BR>allsel<BR>vsel,s,,,4,6<BR>vplo<BR>vsweep,all<BR>allsel<BR>nummrg,all !整理编号 <BR>numcmp,all<BR>allsel<BR>nsel,s,loc,z,0<BR>d,all,all<BR>allsel<BR>nsel,s,loc,z,-7<BR>d,all,all<BR>allsel<BR>nsel,s,loc,x,0<BR>d,all,ux<BR>allsel</P>
<P>/solu<BR>presure=7480370<BR>sfa,21,,pres,presure<BR>allsel<BR>save<BR>solve<BR><BR>请求高手指点??昨天又在5.7版本上试了一下,结果又不同于8.0和9.0</P>
发表于 2006-6-21 16:10 | 显示全部楼层
真是个问题
 楼主| 发表于 2006-6-22 08:30 | 显示全部楼层

谢谢

拜请高手指点!!
发表于 2006-6-22 08:42 | 显示全部楼层
关注中
发表于 2006-6-22 09:21 | 显示全部楼层
<P>是个值得研究的问题。</P>
 楼主| 发表于 2006-6-23 09:18 | 显示全部楼层

新发现

好像在第3步分网时,8.0版本提示正常,9.0版本提示有单元超出了形状极限,并且生成的节点总数与单元总数两个版本也不同。<BR>!选择转向快中部单元划分<BR>allsel<BR>vsel,s,,,2<BR>vplo<BR>esize,0.03<BR>vsweep,all,25,26<BR>
您需要登录后才可以回帖 登录 | 我要加入

本版积分规则

QQ|小黑屋|Archiver|手机版|联系我们|声振论坛

GMT+8, 2024-11-26 10:48 , Processed in 0.095284 second(s), 18 queries , Gzip On.

Powered by Discuz! X3.4

Copyright © 2001-2021, Tencent Cloud.

快速回复 返回顶部 返回列表