Discussion Closed This discussion was created more than 6 months ago and has been closed. To start a new discussion with a link back to this one, click here.

Calculating Time Integral of Torque

Please login with a confirmed email address before reporting spam

Hi everyone!

I am working on optimization of a Vertical Axis Wind Turbine by means of CFD simulation. I have been able to run my model and plot the torque exerted on the blades of the rotor by using a Time Dependent Study (see Figure 1 attached). Next thing I want to do is to compute the average torque over an interval of time so I can calculate de power coefficient of my wind turbine. I have tried to do this by integrating the Torque variable with respect to time using some built-in operators named timeint and timeavg but then my model just take too long to do this calculations and actually it never ends. Does anyone know a better way to do this or know what is my mistake?. I attached some screenshots of my way of getting the Torque values.

Greetings.



4 Replies Last Post Feb 2, 2021, 6:22 p.m. EST

Please login with a confirmed email address before reporting spam

Posted: 3 years ago Oct 27, 2020, 2:47 a.m. EDT

You can use 'with' operator for this. This will give you what you want.

Hope that helps.

You can use 'with' operator for this. This will give you what you want. Hope that helps.

Edgar J. Kaiser Certified Consultant

Please login with a confirmed email address before reporting spam

Posted: 3 years ago Oct 27, 2020, 5:06 a.m. EDT
Updated: 3 years ago Oct 27, 2020, 5:12 a.m. EDT

Eddy,

it is my experience too, that complex postprocessing can be very time consuming. My workaround in some cases is to shift those operations into the solver. The solvers are implemented much more efficient than the post processing. Check this blog contribution by Walter Frei to get some ideas: https://www.comsol.de/blogs/using-the-previous-solution-operator-in-transient-modeling/ Of course you need to modify this according to your needs. Your integrations would then run in additional ODEs making use of the previous solution.

Cheers Edgar

-------------------
Edgar J. Kaiser
emPhys Physical Technology
www.emphys.com
Eddy, it is my experience too, that complex postprocessing can be very time consuming. My workaround in some cases is to shift those operations into the solver. The solvers are implemented much more efficient than the post processing. Check this blog contribution by Walter Frei to get some ideas: https://www.comsol.de/blogs/using-the-previous-solution-operator-in-transient-modeling/ Of course you need to modify this according to your needs. Your integrations would then run in additional ODEs making use of the previous solution. Cheers Edgar

Please login with a confirmed email address before reporting spam

Posted: 3 years ago Nov 18, 2020, 3:57 p.m. EST

Hello.

I have a similar problem, where I would like to calculate the rms (average fluctuating velocity) in my model. I am solving a laminar system but there is "chaotic mixing", so I would like to get an average velocity field () and a rms field (). I was able to get the average velocity field by creating a time average dataset (in postprocessing), but have not been able to get the rms field.

Any help would be greatly appreciated. Thanks

Hello. I have a similar problem, where I would like to calculate the rms (average fluctuating velocity) in my model. I am solving a laminar system but there is "chaotic mixing", so I would like to get an average velocity field (\bar{U}(x,y)) and a rms field (u_{rms}(x,y) = \sqrt{(u^\prime - \bar{U})^2}). I was able to get the average velocity field by creating a time average dataset (in postprocessing), but have not been able to get the rms field. Any help would be greatly appreciated. Thanks

Please login with a confirmed email address before reporting spam

Posted: 3 years ago Feb 2, 2021, 6:22 p.m. EST

Hi! Thank you all for your sugesstions and sorry for my late response. I ended up postprocessing all my data in Matlab, using the LiveLink for Matlab interface and it turned out to be the best way for me to get what I wanted because the postprocessing there is not very time consuming and it's highly customable.

Cheers.

Hi! Thank you all for your sugesstions and sorry for my late response. I ended up postprocessing all my data in Matlab, using the LiveLink for Matlab interface and it turned out to be the best way for me to get what I wanted because the postprocessing there is not very time consuming and it's highly customable. Cheers.

Note that while COMSOL employees may participate in the discussion forum, COMSOL® software users who are on-subscription should submit their questions via the Support Center for a more comprehensive response from the Technical Support team.