I'm trying to understand if I can solve an eigenvalue problem using Dedalus.
Basically, I have an ODE of the form:
(1) L_2(\omega,r) d^2/dr^2 f + L_1(\omega,r) d/dr f + L_0(\omega,r) f = 0
that I want to solve for \omega (the eigenvalues) with r \in ]-\infty,\infty[. The L_0,1,2 are real functions of \omega and r (and other parameters).
I'd like to use the mapping:
(2) r -> H s/(1-s^2)
with H a "spreading" parameter. It converts s \in [-1,1] into r \in ]-\infty,\infty[ with a nice clustering near r=0 (point of interest) if you use the classical chebyshev collocation points.
Do you think I can do this with Dedalus?
The only caveat I see is that there is a problem with the mapping (2) at s=-1,+1 where it is singular. So when transforming (1) in s space, I introduce some terms that are singular at s=-1/+1.
If I were to write my own Cheby solver, I think that these terms could be dealt with by using null BCs (Dirichlet) and removing the problematic lines in the Jacobian*Derivative operator matrix of the eigenvalue problem.
But here, Dedalus (nicely) handles everything. Do you see a simple way to do this manually, i.e. go into Dedalus and tell it to remove the problematic lines in the matricial system?
Thanks!
Louis
Louis
--
You received this message because you are subscribed to the Google Groups "Dedalus Users" group.
To unsubscribe from this group and stop receiving emails from it, send an email to dedalus-users+unsubscribe@googlegroups.com.
To post to this group, send email to dedalu...@googlegroups.com.
To view this discussion on the web visit https://groups.google.com/d/msgid/dedalus-users/0e3d7995-1c64-4c8a-94b1-e94ffacb4113%40googlegroups.com.
For more options, visit https://groups.google.com/d/optout.
To unsubscribe from this group and stop receiving emails from it, send an email to dedalus-user...@googlegroups.com.
To view this discussion on the web visit https://groups.google.com/d/msgid/dedalus-users/CAHqBLzybEfuAQw%3D41ArpQOGzZhJJTRth0n%3DHjkzG1_6muPkVEg%40mail.gmail.com.
Your example notebook is great. Exactly what I was looking for.
However, running it in Jupyter notebook I get the error:
"AttributeError: 'EigenvalueSolver' object has no attribute 'solve_sparse'"
I should say that I haven't used Jupyter very much so I may just be missing something trivial. Maybe it's because my Dedalus install is 1.5 years old?
I just checked the class EigenvalueSolver in dedalus/core/solvers.py and it doesn't seem to include the solve_sparse def.
Thanks,
Louis
--
You received this message because you are subscribed to the Google Groups "Dedalus Users" group.
To unsubscribe from this group and stop receiving emails from it, send an email to dedalus-user...@googlegroups.com.
To post to this group, send email to dedalu...@googlegroups.com.
To view this discussion on the web visit https://groups.google.com/d/msgid/dedalus-users/101f821f-3358-4c7c-b5ef-71ed537dd8ce%40googlegroups.com.
Can I really just do the:
hg pull
hg update
pip3 install -r requirements.txt
python3 setup.py build_ext --inplace
commands and it ll be updated??
To view this discussion on the web visit https://groups.google.com/d/msgid/dedalus-users/5C824E1E-F171-4235-BC07-CDCB41245615%40gmail.com.
To view this discussion on the web visit https://groups.google.com/d/msgid/dedalus-users/CADZXxBh4TQfgjmTtBx8-YNZBZdtT-bKfa70Ds9_spqJJfbrfXA%40mail.gmail.com.
<quantum_oscillator.ipynb>
To view this discussion on the web visit https://groups.google.com/d/msgid/dedalus-users/6381DE0A-FC29-41E4-8D38-8497E01ED57D%40gmail.com.
To view this discussion on the web visit https://groups.google.com/d/msgid/dedalus-users/CADZXxBhNTqUwjmw_J9RGDN9foZHUMU3SkdRiyhQO5%2BBEwizcyw%40mail.gmail.com.
I have a question about the EVP solver.
What's the pencil parameter?
If the EVP is set up with N modes, I'd think that the EVP should be able to spit out N eigenstates.
Do you set a pencil m < N to speed up the calculations? If yes, how do you know that the m eigenstates you're getting are the ones that matters? Is it based on the largest eigenvalues?
Thanks
Louis
--
You received this message because you are subscribed to the Google Groups "Dedalus Users" group.
To unsubscribe from this group and stop receiving emails from it, send an email to dedalus-user...@googlegroups.com.
To post to this group, send email to dedalu...@googlegroups.com.
To view this discussion on the web visit https://groups.google.com/d/msgid/dedalus-users/e46e4ab5-988e-4732-aac3-6940ed84766a%40googlegroups.com.
thanks Jeff!
--
You received this message because you are subscribed to the Google Groups "Dedalus Users" group.
To unsubscribe from this group and stop receiving emails from it, send an email to dedalus-user...@googlegroups.com.
To post to this group, send email to dedalu...@googlegroups.com.
To view this discussion on the web visit https://groups.google.com/d/msgid/dedalus-users/9d44e1bb-82ed-42bd-a967-a6b29ab040c5%40googlegroups.com.
just out of curiosity, in the EVP, is it always better to multiply the eqns such that the denominators are clean? i.e. is there a problem if one of my equations is:
"problem.add_equation(" u + v/r*C + k*w = 0")"
even when C/r is well defined in 0? (r is the independent variable going from -1 to 1)
should I multiply everything by r so that the denominators are ?
--
You received this message because you are subscribed to the Google Groups "Dedalus Users" group.
To unsubscribe from this group and stop receiving emails from it, send an email to dedalus-user...@googlegroups.com.
To post to this group, send email to dedalu...@googlegroups.com.
To view this discussion on the web visit https://groups.google.com/d/msgid/dedalus-users/8c5623fb-a5f0-4ea0-baf4-3810b1c2e4d2%40googlegroups.com.
--
You received this message because you are subscribed to the Google Groups "Dedalus Users" group.
To unsubscribe from this group and stop receiving emails from it, send an email to dedalus-users+unsubscribe@googlegroups.com.
To post to this group, send email to dedalu...@googlegroups.com.
To view this discussion on the web visit https://groups.google.com/d/msgid/dedalus-users/73537e10-2e9d-4d2e-a7db-fa7cdacb1d86%40googlegroups.com.
ok.
I'm trying to reproduce the results in the paper of Mayer and Powell 1992 JFM ("Viscous and inviscid instabilities of a trailing vortex").
I'm getting some errors with the unclean formulation.
I'll let you know if I get it to work with cleaned denominators.
Thanks,
Louis