Modified transport equation difficulties

170 Views Asked by At

I got stuck on part of a question, and I have the feeling there's a simple approach that I'm not thinking of. Here is my work so far:

Ultimately, we would like to solve the PDE

$$u_t+cu_x = u\tag{T}$$ $$u(x,0) = f(x) \tag{IC}$$

With the substitution $u(x,t) = v(x - ct,t)$, we get the modified version

$$v_t = u = v(x-ct,t)$$ $$v(x,0) = f(x)$$

I think that there should be a way of finding a general solution to that first equation $$v_t = v(x-ct,t)$$ Certainly, $Ce^t$ would be a valid solution. However, the general solution must somehow involve an undetermined function of $x$, and I'm not sure how to bring it in to here. My immediate thoughts were to try $\psi(x)e^t$ and $\psi(x-ct)e^t$, but the first never works, and the second will not work for an undetermined differentiable function $\psi$. At this point, I feel as though I've run out of ideas.

Any help or hints are appreciated.

1

There are 1 best solutions below

3
On BEST ANSWER

(I hope my notation is not too horrendous) We used the method of characteristics to solve this equation in my PDEs class. The characteristic equations are given by, $$\frac{\partial{t}}{\partial\tau}=1,\quad\frac{\partial x}{\partial \tau}=c,\quad\text{and}\quad \frac{\partial u}{\partial \tau}=u $$ where $\tau$ is a parameter. I get $t=\tau$, $x=c\tau+x(0)$, and $u=u(x(0),0)e^\tau$. (Here $x(0)$ is just my initial value when our parameter $\tau=0$.) The initial conditions give $$ u=f(x(0))e^t $$ Solving for $x(0)$ gives us that $u(x,t)=f(x-ct)e^t$, which is essentially what you had. Taking partial derivatives, we see this solution satisfies our PDE. Hope that helps!