2
$\begingroup$

Is there a quick smart intuitive way to see that

$$\frac{dx}{F_p} = \frac{dy}{F_q} = \frac{du}{pF_p+qF_q} = - \frac{dp}{F_x+pF_u} = - \frac{dp}{F_y+qF_u}$$

are the characteristic equations for a non-linear pde $F(x,y,u,u_x,u_y)= 0$?

Doesn't even matter how illogical or inapplicable so long as it gives the right result. If I forget them, it'd just take absolutely ages to re-derive them in an exam.

$\endgroup$
1
  • $\begingroup$ The book Physics for mathematicians by Michael Spivak has a nice derivation of the characteristic equations to a first-order PDE. HTH $\endgroup$ Jun 15 '15 at 15:59
1
$\begingroup$

Maybe it's not that intuitive or even easy to remember, but you may want to have a look at the most general case of the Lagrange-Charpit equations:

$$ \frac{\mathrm{d}x_i}{F_{p_i}}=-\frac{\mathrm{d}p_i}{F_{x_i}+F_u \, p_i}=\frac{\mathrm{d}u}{\sum_i \, p_iF_{p_i}}, $$ which are the characteristic equations for the non-linear pde:

$$ F(x_1,x_2, \ldots, p_1,p_2,\ldots, u) = 0, $$

where $p_i = u_{x_i}$ and $x_i$ are the set of indepent variables, whilst $u$ is the unknown.

Cheers!

(Source)

$\endgroup$
2
  • $\begingroup$ I picked the two-variable case for simplicity but if you have a nice intuitive way of remembering the form of those equations I'm all for it. $\endgroup$
    – bolbteppa
    May 9 '14 at 10:43
  • $\begingroup$ Hi there again, @bolbteppa. I'm afraid I have no other more intuitive way to recall these equations. The ones above are those which I usually remember when I need them for a particular 2D or 3D case. $\endgroup$
    – Dmoreno
    May 9 '14 at 10:53

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.