Hi everyone,
I created a 2-bladed VAWT model. The dimension is the same as the turbine used in the experiment.
I have run a 3D transient simulation via sliding mesh. The calculated torque from the simulation is close to the experimental data.
Now, I want to repeat it with the MRF approach. All the settings are the same (mesh, rotational speed, boundary conditions, turbulence model, solver). However, I couldn`t get a good result from the MRF approach. It has far deviated (negative value).
I read from literature, there are only small differences between the torque calculated from MRF and sliding mesh.
I`m just wondering did I miss out on some important steps?
Your settings looks good. Just a few points (1) The rotation direction of the turbine must be in line with the rotation direction set in the software. If the direction is set as opposite of the real situation, you're definitely going o get a negative torque value. (2) I'd also suggest that you take a look at the boundary conditions one more time (including the interface setup).
Goodluck
Hi Victor S. Olawoore ,
Thanks for your suggestions.
I think the axis is correct. In the geometry, +z direction is facing upwards. So, in the frame motion's rotational-axis direction, z = 1, and rotational speed = 20.64 rad/s (positive). Hence, the turbine should be rotating in the counter-clockwise direction. I attached the top view of my geometry.
About the interfaces, I created 3 interfaces (top interface, bottom interface, and side interface). And for all the interfaces walls, I set them to a moving wall with 0 rad/s rotational speed at the same axis as the turbine rotational axis (0, 0, 1). Are these correct? I attached one of the interfaces boundary conditions here.
I would suggest you to enable mesh-motion instead of frame motion for specifying the angular velocity of the rotating domain.
Goodluck@Chun khai Tan
- Badges
- Science topic
- Similar topics
- Computer Science and Engineering
- Computer Graphics and Animation
- 3D