CFD Online Discussion Forums

CFD Online Discussion Forums (http://www.cfd-online.com/Forums/)
-   OpenFOAM (http://www.cfd-online.com/Forums/openfoam/)
-   -   Change axis rotation of MRFZone (http://www.cfd-online.com/Forums/openfoam/94477-change-axis-rotation-mrfzone.html)

trantoan2008 November 17, 2011 01:30

Change axis rotation of MRFZone
 
Hi expert,

Now I am simulating aerodynamic of wind turbine. I should use MRFs function to define rotation speed of it. However, My problem is so difficult due to rotation axis which will be assumed to change dependent with time. I mean rotation axis is not X, Y, or Z axis, conventionally. May I write code to change it with dependent time? Is it possible?
I need some advice to solve above problem.
I really appreciate your help.

Tran

zhaowei November 17, 2011 06:15

what do u mean? u want to change the rotation axis only, or change the whole set-up (including the boundary, rotational axis, and even the grid configuration)?

trantoan2008 November 17, 2011 06:40

Hi Zhao,

Thank your for your quickly response.

Generally, we define X-axis, or Y-axis, or Z-axis for rotation axis (For instance, vector [0 0 1] is defined for Z-rotation axis). However, I want to define axis rotation which depends on time as follow function:
axis rotation = [x(t) y(t) z(t)]
x(t), y(t) and z(t) is function of variable x with dependent time.

Best regards,
Tran

Quote:

Originally Posted by zhaowei (Post 332472)
what do u mean? u want to change the rotation axis only, or change the whole set-up (including the boundary, rotational axis, and even the grid configuration)?


zhaowei November 17, 2011 08:05

Quote:

Originally Posted by trantoan2008 (Post 332478)
Hi Zhao,

Thank your for your quickly response.

Generally, we define X-axis, or Y-axis, or Z-axis for rotation axis (For instance, vector [0 0 1] is defined for Z-rotation axis). However, I want to define axis rotation which depends on time as follow function:
axis rotation = [x(t) y(t) z(t)]
x(t), y(t) and z(t) is function of variable x with dependent time.

Best regards,
Tran

according to your description, i want to know what the "time" mean? what kind of simulation do u want to get? steady state or transient?

i guess u may want something like a pedestal fan, which is a quasi-steady state simulation. that is for each rotation direction, it is a steady state, but generally, the axis is moving. i.e. the movement of the rotation axis is much slower to ensure the flow to reach steady state at each direction. Is that correct? but when u changed the rotation direction, the configuration of grids needs to be adjusted, since the rotation domain should be perpendicular to the rotation axis. this could be very complex, wait for some experts.

trantoan2008 November 17, 2011 08:27

Quote:

Originally Posted by zhaowei (Post 332485)
according to your description, i want to know what the "time" mean? what kind of simulation do u want to get? steady state or transient?

i guess u may want something like a pedestal fan, which is a quasi-steady state simulation. that is for each rotation direction, it is a steady state, but generally, the axis is moving. i.e. the movement of the rotation axis is much slower to ensure the flow to reach steady state at each direction. Is that correct? but when u changed the rotation direction, the configuration of grids needs to be adjusted, since the rotation domain should be perpendicular to the rotation axis. this could be very complex, wait for some experts.

Hi Zhao,

My problem is a transient simulation. When we changed the rotation direction, the configuration of grids needs to be adjusted and the rotation domain should be perpendicular to the rotation axis. I agree with you this point.
For solving rotation zone which must be perpendicular the rotation axis, we may use dynamic mesh technique and GGI technique.

zhaowei November 17, 2011 10:37

wow, sounds not easy. only the ggi has made me frustrating, haha


All times are GMT -4. The time now is 21:50.