Solving symbolic partial differential equation
Mostrar comentarios más antiguos
I am looking to solve a symbolic partial differential equation (PDE), akin to how another symbolic mathematical environment uses the function pdsolve().
Is there an equivalent in the MATLAB symbolic toolbox? I have found dsolve, however am not sure how much of the behaviour would be equivalent.
So far I have only received the error message "Indeterminates must be functions." I am thinking this must be due to how I have set up
f(t1, t2, t3, p2, p3, p4)
In my minimum working example:
%%MWE
syms t1 t2 t3 p2 p3 p4
pars = [t1 t2 t3 p2 p3 p4];
d = 1;
MM = length(pars);
alphapre = [0; 0; -t3/p4; 0; 0; 1];
syms f(t1, t2, t3, p2, p3, p4)
alpha = sym(zeros(d, MM));
PDE = sym(zeros(d,1));
for mm = 1:d
alpha(mm, :) = alphapre(:,mm).';
PDE(mm) = sum(alpha(mm,:).*jacobian(f,pars));
end
dsolve(PDE == 0);
which results in a PDE of the form
d
t3 --- f(t1, t2, t3, p2, p3, p4)
d dt3
--- f(t1, t2, t3, p2, p3, p4) - -------------------------------- == 0
dp4 p4
The answer to this example should be
t3*t4
Have I made a stupid mistake? Or is it just that I should use something different for this functionality?
2 comentarios
Torsten
el 27 de Feb. de 2018
"dsolve" solves ODEs, not PDEs.
There is no official MATLAB tool that symbolically solves PDEs.
Best wishes
Torsten.
Ben Holmes
el 27 de Feb. de 2018
Respuestas (0)
Categorías
Más información sobre Eigenvalue Problems en Centro de ayuda y File Exchange.
Productos
Community Treasure Hunt
Find the treasures in MATLAB Central and discover how the community can help you!
Start Hunting!