a strange problem when trying to perform a three-dimensional microscopic modeling of masonry structure.

Post Reply
xgx
Posts: 40
Joined: Sun Sep 24, 2023 2:39 pm

a strange problem when trying to perform a three-dimensional microscopic modeling of masonry structure.

Post by xgx » Sat Oct 28, 2023 11:41 am

Dear STKO team, I encountered a strange problem when trying to perform a three-dimensional microscopic modeling of masonry structure. However, when I conducted a gravity analysis on one wall alone, the analysis proceeded normally and there were no issues. However, when I merged several walls to create a masonry house, I encountered the error in the following image. I really don't understand why. The boundary conditions are set as shown in the figure below, and I don't think there is any problem. This question has been bothering me for several days, and I would greatly appreciate it if you could answer it for me.
Attachments
不收敛2.png
不收敛2.png (663.32 KiB) Viewed 4867 times
不收敛1.png
不收敛1.png (49.69 KiB) Viewed 4867 times

STKO Team
Posts: 3066
Joined: Tue Oct 29, 2019 8:45 am

Re: a strange problem when trying to perform a three-dimensional microscopic modeling of masonry structure.

Post by STKO Team » Mon Oct 30, 2023 9:13 am

If you see the line "Warning: numeric analysis returns -1" it means that your model is singular.
From what I can see from the image, your slab is not attached to the walls

xgx
Posts: 40
Joined: Sun Sep 24, 2023 2:39 pm

Re: a strange problem when trying to perform a three-dimensional microscopic modeling of masonry structure.

Post by xgx » Mon Oct 30, 2023 10:02 am

STKO Team wrote:
Mon Oct 30, 2023 9:13 am
If you see the line "Warning: numeric analysis returns -1" it means that your model is singular.
From what I can see from the image, your slab is not attached to the walls
Dear STKO team:
Thanks for your reply, but it seems that you misunderstood what I want to express. In fact, I have merged the slab and wall, but I accidentally set the shader color of the slab to red, which caused you to misunderstand. Even if I removed the slab, only four walls remained, merged them, fixed the bottom, and ran the analysis, the program still reported this error. This really troubles me. I have attached my SCD file. Could you please help me check it.
Attachments
cyclic test of masonry structure3_r6.zip
(10.59 MiB) Downloaded 241 times
不收敛3.png
不收敛3.png (562.61 KiB) Viewed 4846 times

xgx
Posts: 40
Joined: Sun Sep 24, 2023 2:39 pm

Re: a strange problem when trying to perform a three-dimensional microscopic modeling of masonry structure.

Post by xgx » Mon Oct 30, 2023 10:22 am

xgx wrote:
Mon Oct 30, 2023 10:02 am
STKO Team wrote:
Mon Oct 30, 2023 9:13 am
If you see the line "Warning: numeric analysis returns -1" it means that your model is singular.
From what I can see from the image, your slab is not attached to the walls
Dear STKO team:
Thanks for your reply, but it seems that you misunderstood what I want to express. In fact, I have merged the slab and wall, but I accidentally set the shader color of the slab to red, which caused you to misunderstand. Even if I removed the slab, only four walls remained, merged them, fixed the bottom, and ran the analysis, the program still reported this error. This really troubles me. I have attached my SCD file. Could you please help me check it.
At present, what I can confirm is that all the walls have been merged into a masonry structure. I fixed the bottom of this structure and conducted gravity analysis, but still encountered the error of singularity in the stiffness matrix. Because of this, I really don't understand why. I sincerely desire for your reply.

STKO Team
Posts: 3066
Joined: Tue Oct 29, 2019 8:45 am

Re: a strange problem when trying to perform a three-dimensional microscopic modeling of masonry structure.

Post by STKO Team » Mon Nov 06, 2023 11:58 am

It runs in my case.
Probably your model is too large (too many DOFs) for a single processor.
Try using parallel analysis with OpenSeesMP

xgx
Posts: 40
Joined: Sun Sep 24, 2023 2:39 pm

Re: a strange problem when trying to perform a three-dimensional microscopic modeling of masonry structure.

Post by xgx » Mon Nov 06, 2023 12:29 pm

STKO Team wrote:
Mon Nov 06, 2023 11:58 am
It runs in my case.
Probably your model is too large (too many DOFs) for a single processor.
Try using parallel analysis with OpenSeesMP
Thank you for your advice, however, I tried to do as you said, but it doesn't work.

xgx
Posts: 40
Joined: Sun Sep 24, 2023 2:39 pm

Re: a strange problem when trying to perform a three-dimensional microscopic modeling of masonry structure.

Post by xgx » Mon Nov 06, 2023 12:31 pm

xgx wrote:
Mon Nov 06, 2023 12:29 pm
STKO Team wrote:
Mon Nov 06, 2023 11:58 am
It runs in my case.
Probably your model is too large (too many DOFs) for a single processor.
Try using parallel analysis with OpenSeesMP
Thank you for your advice, however, I tried to do as you said, but it doesn't work.
So, the reason for this issue is actually due to insufficient performance of my computer?

xgx
Posts: 40
Joined: Sun Sep 24, 2023 2:39 pm

Re: a strange problem when trying to perform a three-dimensional microscopic modeling of masonry structure.

Post by xgx » Mon Nov 06, 2023 12:48 pm

xgx wrote:
Mon Nov 06, 2023 12:31 pm
xgx wrote:
Mon Nov 06, 2023 12:29 pm
STKO Team wrote:
Mon Nov 06, 2023 11:58 am
It runs in my case.
Probably your model is too large (too many DOFs) for a single processor.
Try using parallel analysis with OpenSeesMP
Thank you for your advice, however, I tried to do as you said, but it doesn't work.
So, the reason for this issue is actually due to insufficient performance of my computer?
Thank you very much, your suggestion inspired me. After increasing the grid and reducing the number of grids, it can run normally.


Post Reply