asdeasoft.net

I am running this code to get analysis report for first 200 modes. However when I am running I am getting error like

missing operand at _@_
in expression "...0000000000000000000 * _@_"

Please let me know reason for the above error
Thank you

Dear akshaysomkuwar123,

In tcl scripting the error you mention normally occurs when in a braced argument there is no operator between two operands.
This can happen if you try to substitute a variable containing an operand inside a braced argument, which by itself does not allow substitution. Or when an expression to evaluate is passed in as a variable or parameter.

But in reference to the script originally posted in this thread, as far as I know it still works properly.
Could you post your .scd file zipped with the custom tcl script you are trying to run?

In addition, you can check out this other thread on modal analysis:
viewtopic.php?f=33&t=1823&p=2838&hilit=modal#p2886

and the webinar on the new modalProperties command added in the new version of STKO and included in OpenSees 3.2.2:
https://www.youtube.com/watch?v=vUXKVm5_rn0

We await from your feedback.
Francesca

Dear Francesca

Please find the tcl script for the modal on which I am working.

Yes the script posted here is working properly when I am asking for small mode numbers (20-50) but giving the above error which I have posted for large mode numbers.

Regards,
Akshay S

We had a look at your file.
It does not work in the modal script because, from about mode 150, you are having very strange eigenvectors.. and at mode 169 the eigenvector components reach values as high as 1.0e14.
So, when the modal script tries to get the eigenvector in TCL, it receives a wrong string and cannot perform the operation.

Try to plot the modal shapes with the MPCO recorder and you will see what happens from about mode 150.

Try to review your model, and make sure you really need all those modes.

Thank you very much for reviewing the model.
I will try to plot the modes and understand what is happening.
I am trying to replicate the model given in article, the author has considered 200 modes.

Dear akshaysomkuwar123,

let us know how it goes. If you wish, you can also post the article reference if you'll need to ask more questions on it.

Enjoy your modeling

Francesca

Dear akshaysomkuwar123,
I have an additional note to what we suggested you before.
You should check if in the article you are referencing, the mass were modeled only as translational masses, without the rotational components.

In your tcl script you did not assign any rotational masses, the highest modes you are researching could be rotational ones, with very high frequencies. Without even a very small value of rotational masses applied, opensees will not be able to capture them, as it can encounter numerical errors with iterative processes on sparse matrices.

Try to add a very small value, like two or three order of magnitude smaller than the masses you applied in x,y,z and see if you get better results.

A tip for the fast substitution of the mass value is to use the regular expression available in Notepad++
https://datacadamia.com/ide/notepad/replace

Enjoy!
Francesca

Dear Francesca

Thank you for the suggestions.
Please find the attached article in the thread. The author has done analysis in the Abaqus Software, so it is difficult to get an idea how to assign masses and get similar results.
Currently I am trying to figure out problem with too large eigenvector as you pointed out in previous threads, Please let me know any suggestions over it what it might be causing such problem.

Thank you,

Regards,
Akshay S

Dear akshaysomkuwar123,

in the article you sent, at section 3.4 it states: The ball centres of the rigid bodies were assigned a mass of 64 kg and rotary inertia of 0.378 kgm2.

So the rotational masses are considered in the modelling, and they are the reason why you have those problems in the higher modes, include them in the appropriate units and you'll see that it will solve your problem.

Francesca

Dear Francesca

Sorry I did not mention earlier about the model, please look at section 4.0 Finite element model of prototype single-layer reticulated dome.
Please let me know any suggestions.

Thank you,

Regards,
Akshay S