-
-
Notifications
You must be signed in to change notification settings - Fork 224
Generalize symbolic indexing of ODE/DDE to multivariate variables #3738
New issue
Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.
By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.
Already on GitHub? Sign in to your account
base: master
Are you sure you want to change the base?
Conversation
Any input on this? @ChrisRackauckas / @AayushSabharwal |
I'm a little on the fence about this. Is it just for UX? |
That's not an ODE/DDE, so it should fail. It's not solvable with these techniques so it should fail at construction. Maybe we should throw a better error but an ODE solver cannot be directly applied to a PDE without proper handling of the other independent variable. |
I totally see where you are coming from, but I think it is slightly more nuanced.
It can if the PDE is an ODE. In my application this is precisely a PDE that reduces to one ODE in
The dependence on It is 100% intentional and with the understanding that my ODEs are reduced PDEs that I write |
That would require infinitely many ODE solves then.
That is very different. For that you have an array variable for a system of equations. |
It sure does 😅 The ODEs are big and stiff, too. I use
The Fourier transform is in the infinite-space limit, so |
This is how you're discretizing the PDE. The ODE solver doesn't have ways to discretize a PDE |
Yes, sure. |
There are a lot of assumptions that have to be true for this to work out. The problem isn't that your case has a trivial mapping to many ODEs, it's that many other cases do not. You could have a term For this one, not only do you have that the equations are independent across But that doesn't mean MTK doesn't cover this. You'd just phrase it as a PDESystem and then we'd need to add a If you do it like that, then you'd also get the PDE plotting, so you could plot slices of |
This is one possible fix to #3737.$x(t-\tau)$ to permit multivariate $x(t-\tau, a_2, a_3, \ldots)$ .
It generalizes a line that assumed DDE variables are like
Indexing ODEs with multivariate variables works "for free" when the DDE line no longer errors.
Checklist
contributor guidelines, in particular the SciML Style Guide and
COLPRAC.