'VAWA_Harmonic_Analysis' simulation project by Iria


#1

I created a new simulation project called 'VAWA_Harmonic_Analysis':

Harmonic analysis of the VAWA with inputs from the airflow analysis and the frequency analysis


More of my public projects can be found here.


#2

Thank you very much @afischer!
It would be great if you provide me with the update. I need the results of the frequency analysis (the eigenfrequencies) for the harmonic analysis.
I have a lot of questions about the settings of the LOADS (boundary conditions) in the simulation of harmonic analysis. I want to consider as INPUT for the harmonic analysis all the loads applied during the simulation of the airflow around the wind turbine. But I don´t know exactly which are all the loads that I have to consider. Can you help me with this? I´m thinking on considering:

  • the centrifugal force
  • the atmospheric pressure
  • drag force
  • lift force
  • moment force
  • pressure force and moment ?
  • the porous and viscous force and moment are not so important to be considered ?
    It is right?
    Thank you very much @afischer! I hope you can help me with this!

#3

Hi @afischer!
I have a problem with the project “VAWA_Harmonic_Analysis”. I do not know what´s the problem with my csv-table of the eigenfrequencies.
Thank you very much!


#4

Hi @Iria, I am not very familiar with your model but maybe I can jump in here and help you out in regards to the harmonic analysis.
Generally loads can be input into a harmonic analysis via functions or tables. With this feature it is possible to approximately apply the fluid pressure field as a load for the harmonic analysis (something like the total fluid pressure on a surface, or a linearly varying fluid pressure), but until now we can not directly extract the fluid forces and use them as a direct input for a mechanical calculation.
I am not entirely sure which forces would be the most relevant maybe @gholami could help us out here?

With regards to the centrifugal load, I would like to mention that it is currently not used to compute a gyroscopic matrix and thus the resonance frequency will not depend on the rotation speed, only the resulting displacements and stresses. We are working on including this as well.

Best,
Richard


#5

@Iria, your table is missing one column.
With the following settings:

<img src="/forum/uploads/default/original/2X/7/7bd50dc2f482e4f8a49e985cc6848e6f30a26a70.png" width="441 height=“500”>

the requested format of the table would be:

freq1, delta_freq1
freq2, delta_freq2

This means from your Start frequency until freq1 every delta_freq1 Hz will be computed
and from freq1 until freq2 every delta_freq2 Hz will be computed.

So for the above settings and this table

the computed frequencies would be:

1.0, 1.2, 1.4, 1.6, 1.8, 2.0, 3.0, 4.0, 5.0

I hope the explanation makes sense.
Best,
Richard


#6

Hi @rszoeke!
ok , I think I understand, but then if it is as you say, the table should be:
2, 0.2
5, 1
From start frequency until 2 will be every 0.2 Hz computed and from 2 till 5 will be every 1 Hz computed. It is right?


#7

You are right! That was a mistake, I’ll correct it.


#8

Thank you @rszoeke!
I hope that @gholami or someone else could help me with this. I have just try to run a harmonic analysis applying:

  • centrifugal load
  • atmospheric pressure
  • drag force
  • lift force
    but it did not work. To calculate the drag and lift forces I used the drag and lift coefficients of the airflow simulation and these equations:
  • Fd=0.5CdρAV^2 [N]
  • Fl=0.5ClρAV^2 [N]
    Then, depending on the drag and lift direction, I calculate the x and y components of the drag and lift forces and I used these as boundary conditions. But I´m not sure of this is correct.
    You said below, that I can apply the fluid pressure field as a load for the harmonic analysis. How can I do that?
    My objective is to simulate and study the vibrations on the wind turbine produced while rotating. I hope it is possible with simscale.
    Thank you for everything!

#9

@Iria, I guess you mixed the column for freq and delta_freq:

and your table looks like:

so you get the error message that the Timestep bigger than the given simulation interval:

Timestep might be not exactly correct here, since it is rather a frequency step, but the solution is the same.


#10

Hi @rszoeke!
I think I solve the problem with the csv-table but now I became the error: “Matrix not factorable”. I eliminate all the loads that I defined, I do not have any constraints. I do not know what is the problem now.
Best
Iria


#11

Hi @Iria, I guess you should add a constraint to your model, presumably the faces at the center which would connect to the shaft would be a good choice for a fixed support.
For high load frequencies it might not be needed to have a constraint but for very low frequencies you might get into troubles by numerical singularities like the mentioned “Matrix not factorable”=“Matrix not inversible”.

Additionally you should not add constant loads like the atmospheric pressure, since for a harmonic analysis every load (besides pretension loads) are applied harmonically, this means as sinusoidal function with the defined value as magnitude.

Pretension loads are currently under development and will be available soon. In your case I would guess that they would not have an influence anyways, besides gravity perhaps.


#12

Hi @Iria, regarding which forces to take into account: your treatment of drag and lift forces are correct, and I would include them as you have. I don’t think inclusion of atmospheric pressure is necessary though. You’ve mentioned centrifugal load as well; how do you take it into account?