Solutions to PDE: Understanding and Simplifying the Process

  • Context: Graduate 
  • Thread starter Thread starter AntSC
  • Start date Start date
  • Tags Tags
    Pde
Join the discussion
Ask a follow-up here, or get your own question answered by working scientists, mathematicians and engineers — people, not an autocomplete.
Real named experts · corrections over time · the nuance an AI answer skips
4 replies · 3K views
AntSC
Messages
65
Reaction score
3
I have seen a couple of solutions to this PDE -

[itex]\frac{\partial x}{\partial u}=\frac{x}{\sqrt{1+y^{2}}}[/itex]

One is -

[itex]u=\ln \left | y+\sqrt{1+y^{2}} \right |+f\left ( x \right )[/itex]

I have no idea how this is arrived at or if it's correct. This is what i want to know.

The solution I've checked out makes the substitution of [itex]y=\sinh \theta[/itex] giving -

[itex]u=x\ln \left | \cosh \theta \right | + f\left ( x \right )[/itex]

which is where I'm a bit stuck as substituting in [itex]\theta =\sinh^{-1} y[/itex] gives a [itex]\cosh \sinh^{-1} y[/itex] term that i don't know how to simpify.

Any help with both of these would be appreciated
 
Physics news on Phys.org
AntSC said:
I have seen a couple of solutions to this PDE -

[itex]\frac{\partial x}{\partial u}=\frac{x}{\sqrt{1+y^{2}}}[/itex]
I am confused by your notation. You have x differentiated with respect to u but on the right both x and y. If you really mean what you have, it is not a "partial" differential equation at all. We are thinking of x as a function of u, with y a fixed parameter. Further that is a separable equation: [itex]dx/du= x\sqrt{1+ y^2}[/itex] gives
[tex]\frac{dx}{x}= \frac{du}{\sqrt{1+ y^2}}[/tex]
[tex]ln|x|= \frac{u}{\sqrt{1+ y^2}}+ c[/tex]
[tex]x= Ce^{\frac{u}{\sqrt{1+ y^2}}}[/tex]
where [itex]C= e^c[/itex].

One is -

[itex]u=\ln \left | y+\sqrt{1+y^{2}} \right |+f\left ( x \right )[/itex]

I have no idea how this is arrived at or if it's correct. This is what i want to know.

The solution I've checked out makes the substitution of [itex]y=\sinh \theta[/itex] giving -

[itex]u=x\ln \left | \cosh \theta \right | + f\left ( x \right )[/itex]

which is where I'm a bit stuck as substituting in [itex]\theta =\sinh^{-1} y[/itex] gives a [itex]\cosh \sinh^{-1} y[/itex] term that i don't know how to simpify.

Any help with both of these would be appreciated
Your original equation treats x as a function of u but your answer gives u as a function of x. Which is intended? if the latter, then
[tex]\frac{du}{dx}= \frac{\sqrt{1+ y^2}}{x}[/tex]
[tex]\frac{du}{\sqrt{1+ y^2}}= \frac{dx}{x}[/tex]
[tex]\frac{u}{\sqrt{1+ y^2}}= ln|x|+ C[/tex]
[tex]u= (ln|x|+ C)\sqrt{1+y^2}[/tex]
essentially just solving the previous equation for u.
 
Sorry, I've written the dependent and independent variables the wrong way round and got the independent variable wrong. It's been a long week!

It should be -

[itex]\frac{\partial u}{\partial y}=\frac{x}{\sqrt{1+y^{2}}}[/itex]

for a function [itex]u\left ( x,y,z \right )[/itex]

There will be a constant term in the form [itex]f\left ( x,z \right )[/itex] in addition to whatever the result of the integration yields (which i can't work out).
 
Last edited:
I've managed to get it sussed. Thanks everyone for the help. Even if it was for me to see that I've written the question down wrong! :smile: