Solving a Partial Differential Equation with the Characteristic Method

Click For Summary
The discussion focuses on solving a partial differential equation using the characteristic method, specifically addressing the equations dx/dt = 1, dy/dt = x, and du/dt = (y - 1/2x^2)^2. The initial conditions provided are x(0,s) = 0, y(0,s) = s, and u(0,s) = e^s. The user successfully derives x(t,s) = t and y(t,s) = 1/2t^2 + s but struggles with the du/dt equation, initially substituting variables incorrectly. Clarification reveals that du/dt can be integrated directly as s^2, leading to the function u(s,t) = s^2 * t, although it must satisfy the initial condition. The conversation emphasizes the importance of correctly applying initial conditions in solving PDEs.
The Head
Messages
137
Reaction score
2
Homework Statement
u_x + x*u_y = (y-1/2x^2)^2, u(0,y) = e^y
Relevant Equations
du/dt = du/dx*dx/dt + du/dy*dy/dt
dx/dt =1, x(0,s)=0, dy/dt=x, y(0,s) = s, du/dt=(y-1/2x^2)^2, u(0,s)=e^s

I did well at the beginning to get x(t,s) =t and y(t,s)=1/2t^2 + s, but got stuck with the du/dt part.

You can sub in x=t and y=1/2t^2 +s for x and y to get du/dt = s^2. But that's still three variables, and I can't see any obvious substitutions.

This is a regular diffy Q course, so I don't have a lot of knowledge of advanced techniques, and this was just thrown in, so I'm a little unsure how to proceed.
 
Physics news on Phys.org
Also, I tried u=e^s => s=ln(u), so du/dt = (ln(u))^2, but that gives me an Ei function, so I feel like that can't be correct.
 
The Head said:
You can sub in x=t and y=1/2t^2 +s for x and y to get du/dt = s^2.

The expression
$$
\frac{\partial u}{\partial t} = s^2
$$
can be directly integrated, ##s## is the other coordinate - not an unknown to be solved for.

Edit: Missed the square in the original PDE.
 
Last edited:
Orodruin said:
The expression
$$
\frac{\partial u}{\partial t} = s^2
$$
can be directly integrated, ##s## is the other coordinate - not an unknown to be solved for.

Edit: Missed the square in the original PDE.

Ohh, OK gotcha. So it can just be s^2 * t = u(s,t)? Thanks so much for your reply.
 
The Head said:
Ohh, OK gotcha. So it can just be s^2 * t = u(s,t)? Thanks so much for your reply.
No, but almost. Your function does not satisfy the initial condition.
 
Orodruin said:
No, but almost. Your function does not satisfy the initial condition.
Right, of course! Thank you.
 
First, I tried to show that ##f_n## converges uniformly on ##[0,2\pi]##, which is true since ##f_n \rightarrow 0## for ##n \rightarrow \infty## and ##\sigma_n=\mathrm{sup}\left| \frac{\sin\left(\frac{n^2}{n+\frac 15}x\right)}{n^{x^2-3x+3}} \right| \leq \frac{1}{|n^{x^2-3x+3}|} \leq \frac{1}{n^{\frac 34}}\rightarrow 0##. I can't use neither Leibnitz's test nor Abel's test. For Dirichlet's test I would need to show, that ##\sin\left(\frac{n^2}{n+\frac 15}x \right)## has partialy bounded sums...