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.
Injection of velocity fluctuations inside the computational domain
Posted Nov 14, 2016, 10:09 a.m. EST Acoustics & Vibrations, Modeling Tools & Definitions, Parameters, Variables, & Functions Version 5.2a 0 Replies
Please login with a confirmed email address before reporting spam
Currently getting started with COMSOL, I am trying to carry out a simple simulation to learn how to use the software, but I seem to get stuck somewhere in the process.
The present problem basically consists in injecting velocity fluctuations inside the computational domain and observe their convection by a simple uniform background uniform mean flow. The fluctuations of velocity should be introduced at the upstream boundary and shouldn’t generate any pressure fluctuations (meaning the velocity fluctuations should be divergence-free).
To solve this 2D problem, the Linearized Euler equations are resolved ("Linearized Euler, Transient", time-dependent study, in the Aeroacoustics tab of the Acoustics Module) The following parameters are used:
Background mean flow: u0 = 170m/s in the x-direction
Boundary conditions:
- upstream: “Prescribed Acoustic Field” with a prescribed single-frequency vertical velocity fluctuation in the y-direction of the shape u’ = 2.4*cos(k*x-omega*t) with k = 0.2 rad/m and omega = k*u0.
- “Asymptotic Far-Field Radiation” for the three other boundaries
Meshing: regular Cartesian mesh with 101 x 101 points and a regular spacing of 1m.
Solver: time step = 0.003 s
Initial values: velocity field imposed in the entire domain to avoid any discontinuity at the boundary.
After computing, we can observe that the velocity already diverges after few time steps only, which shouldn't be the case. Some “parasites” can also already be seen at the corners of the upstream boundary at t = 0, which I can’t explain according to the initial values I’ve imposed.
I have tried working on the time solver to verify that the CFL condition was satisfied, as well as on the meshing to assure a good spatial discretization (at least 30 points per wavelength). I am also wondering whether the boundary condition chosen to impose the velocity fluctuation upstream is maybe inducing some numerical reflection. However, I don’t know what other boundary condition could be chosen to impose these velocity fluctuations, without generating any pressure fluctuations. Is it possible in COMSOL to modify the equation of existing boundary conditions / implement our own boundary conditions?
I have attached the simulation in this post to have a better idea of what the problem is.
Would anyone have an idea to try out?
Thank you for your help.
Marine
Attachments:
Hello Marine Cannard
Your Discussion has gone 30 days without a reply. If you still need help with COMSOL and have an on-subscription license, please visit our Support Center for help.
If you do not hold an on-subscription license, you may find an answer in another Discussion or in the Knowledge Base.