Solving xu_t + uu_x = 0: Need Help with Change of Variables

Click For Summary

Discussion Overview

The discussion revolves around solving the partial differential equation $$xu_t + uu_x = 0$$ with the initial condition $$u(x, 0) = x$$. Participants explore the implications of a suggested change of variables, specifically transforming $$x$$ to $$x^2$$, and whether this is necessary for solving the problem using the method of characteristics.

Discussion Character

  • Exploratory
  • Technical explanation
  • Debate/contested
  • Mathematical reasoning

Main Points Raised

  • One participant expresses uncertainty about how to apply the suggested change of variables and questions its necessity, considering the method of characteristics as a viable approach.
  • Another participant suggests that the hint may be relevant if the problem were simplified to $$u_t + uu_x = 0$$, indicating familiarity with the method of characteristics.
  • A third participant elaborates on the general procedure for solving nonlinear PDEs using characteristics, providing a detailed breakdown of the system of ordinary differential equations involved.
  • This participant also derives expressions for $$t$$, $$v$$, and $$x$$, ultimately leading to a proposed solution for $$u(x, t)$$.
  • A later reply acknowledges the confusion regarding the use of the same variable $$x$$ in the change of coordinates and states that they successfully solved the problem using the method of characteristics without the coordinate transformation.

Areas of Agreement / Disagreement

Participants do not reach a consensus on the necessity of the change of variables. While some explore its implications, others successfully solve the problem without it, indicating multiple approaches to the solution.

Contextual Notes

There are unresolved aspects regarding the effectiveness of the suggested change of variables and its impact on the solution process. The discussion reflects varying levels of familiarity with the method of characteristics and the specific problem at hand.

ryo0071
Messages
12
Reaction score
0
Okay so I am working on this problem:

Solve $$xu_t + uu_x = 0$$ with $$u(x, 0) = x.$$(Hint: Change variables $$x \rightarrow x^2$$.)

However, I am not sure how to use the change of variables hint that is given or why it is needed. My thinking is that I could just use the method of characteristic as normal to get the solution. Any help is appreciated.
 
Physics news on Phys.org
Yes, I'm not sure why the hint is even suggested unless maybe you solved

$u_t + u u _x = 0,\;\; u(x,0)=f(x)$.

already. The method of characteristic will work fine. Are you familiar with this method?
 
ryo0071 said:
Okay so I am working on this problem:

Solve $$xu_t + uu_x = 0$$ with $$u(x, 0) = x.$$(Hint: Change variables $$x \rightarrow x^2$$.)

However, I am not sure how to use the change of variables hint that is given or why it is needed. My thinking is that I could just use the method of characteristic as normal to get the solution. Any help is appreciated.

The general solving procedure of a nonlinear PDE of the type... $\displaystyle u_{t} + \varphi(u,x)\ u_{x} = f(x,t),\ u(x,0)= g(x)\ (1)$

... is the following. First we write the system of three ODE...

$\displaystyle \frac{d t}{d s}=1,\ t(0)=0$

$\displaystyle \frac{d x}{d s}= \varphi(u,s),\ x(0)= \xi$

$\displaystyle \frac{d v}{d s} = f(x,t),\ v(0)=g(\xi)\ (2)$... then we solve (2) obtaining $\displaystyle t=t(s, \xi),\ x=x(s,\xi), v=v(s, \xi)$ and then $\displaystyle s=s(x,\ t), \xi=\xi(x,\ t)$. The solution of (1) is...

$\displaystyle u(x,\ t) = v \{s(x,\ t),\ \xi(x,\ t)\}\ (3)$

In Your case is $\displaystyle \varphi (u,\ x) = \frac{u}{x},\ f(x,t) = 0,\ g(x)=x$ so that the (2) becomes...

$\displaystyle \frac{d t}{d s}=1,\ t(0)=0$

$\displaystyle \frac{d x}{d s}= \frac{v}{s},\ x(0)= \xi$

$\displaystyle \frac{d v}{d s} = 0,\ v(0)=\xi\ (4)$

With a little of patience You solve (4) obtaining...

$\displaystyle t=s$

$\displaystyle v=\xi$

$\displaystyle x=\sqrt{2 s \xi + \xi^{2}}\ (5)$

If You invert (5) obtain...

$\displaystyle s=t$

$\displaystyle \xi = -t + \sqrt{t^{2} + x^{2}}\ (6)$

... and (6) permits You to write...

$\displaystyle u(x,\ t)= -t + \sqrt{t^{2} + x^{2}}\ (7)$

Kind regards

$\chi$ $\sigma$
 
Thank you both for the help. I was trying to use the hint but what was confusing me was that they used the same variable x for both parts of the change of coordinates (rather than doing something like $$x \rightarrow \eta^2$$). Anyway, I was able to solve it by method of characteristics without using the coordinate transform. Again, thank you both for the assistance.
 

Similar threads

  • · Replies 14 ·
Replies
14
Views
2K
Replies
8
Views
2K
  • · Replies 15 ·
Replies
15
Views
2K
  • · Replies 2 ·
Replies
2
Views
3K
  • · Replies 1 ·
Replies
1
Views
3K
  • · Replies 1 ·
Replies
1
Views
2K
  • · Replies 6 ·
Replies
6
Views
2K
Replies
2
Views
1K
Replies
4
Views
2K
  • · Replies 3 ·
Replies
3
Views
4K