trying to solve this first order pde
\[u_t+xtu_x=0\]\[u(x,0)=f(x)\]\[-\infty
my prof really skimmed over this, I have no idea where to start here. I can't find any characteristic equations...
\[\dot{x}=xt\]\[\ln x=\frac{t^2}{2}+C\]\[x(t)=Ce^{t^2/2}\]
How about Separation of Variables?
I don't see how SOV would work?
Try now. Let me see your work,
\[\frac{u_t}{t}=-xu_x\]
... and i'm stuck haha.
That's a simple DE to solve.
And I meant Separation of Variables by assuming U is a product of x and t functions.
Meaning, let U = R(x)*G(t).
oh. you mean like every other SOV in this course? oops. just a second.
\[u(x,t)=R(x)G(t)\]\[R(x)G'(t)=-xtR'(x)G(t)\]\[\frac{G'(t)}{tG(t)}=-\frac{xR'(x)}{x}=\text{constant}\equiv C\] \[G'(t)=CtG(t)\]\[G(t)=Ce^{kt^2/2}\] \[\frac{R'(t)}{R(t)}=-\frac{k}{x}\]\[\ln R(t)=-k \ln x+C\]\[R(t)=Cx^{-k}\] \[u(x,t)=C\frac{e^{kt^2/2}}{x^k}\]
I didn't go through your work, but it looks like you get the point.
oops. I messed up copying it.
This is the first method you should always attempt.
I still don't have a solution that works. not even close.
well I do. but I have no idea how to "get" it.
What's the problem?
Your answer will be in terms of f(x).
if I could get to this\[u(x,t)=\frac{t^2}{2}-\ln x\]I would be stoked.
Is that the final answer?
I don't know. but that satisfies the pde. just from inspection
Dude/Dudette, there will be many solutions to the PDE.
Use the initial condition given.
u(x,0)=f(x) is also true with the solution I have?
I dunno honestly I am getting nowhere with separation of variables.
No, because nobody said what f(x) is. Your answer should be in terms of f(x). Said that already.
|dw:1396235748660:dw|
Join our real-time social learning platform and learn together with your friends!