Solve the heat equation
\[\alpha^2\frac{\partial^2u}{\partial x^2}=\frac{\partial u}{\partial t}\], 0
I am sorry, I don't know
Can you help me check mine? I just imitate what my prof do without understanding a word
\[\alpha^2\frac{\partial^2u}{\partial x^2}=\frac{\partial u}{\partial t}\]
yes
What do you mean, "check" did you already solve it?
I told you, I just imitate my prof's stuff, I didn't understand
I guess you are using Fourier series right..
I really don't know
I see neumann conditions. So shouldn't it be fourier cosine series?
no no, wait you do have fourier sine series! Sorry, you're right!
I am looking at your 1, but i gotta step out for a bit.
your boundaries should be like this, do you think you have typo\[U_{x}(0,t)=0 \\U_{x}(L,t)=0\\ U_{x}(x,0)=x(L-x)\]
nope, there is no x as sub in U (0,t) or U(L,t)
ok this is different problem..
I got Cn using by mathematica..
It cannot have sin there, right? because sin ... pi =0
yes
when we simplify we get \[\frac{4L^2}{n^3\pi^3}-\frac{4L^2\cos n \pi}{n^3\pi^3}\]
you forgot 2 when you calculate c_1
I don't get why at out(4) , numerator says 2L^2 (2+ (-2+ n^2 pi^2...) they don't cancel out 2 and -2 inside the bracket?
yeah, you are right, thank you
why my denominator just n^2pi^2 , theirs is n^3pi^3:(
your C_1 is right, let me check C_2
I see my mistake at denominator from page 3, the second term should be n^3 pi^3 at denominator, ha!!
that's right!!
\[C_{1}=-\frac{2L^2\cos n \pi}{n \pi}\\C_{2}=\frac{4L^2\cos n \pi}{n \pi}-\frac{4L^2\cos n \pi}{n^3 \pi^3}\]
c2 has another term without cos term
nope, I combine C_1 and C_2 , cancel out some like term to get \[C_n= -\dfrac{4L^2 cos(n\pi)}{n^3\pi^3}-\dfrac{2L^2 sin(n\pi)}{n^2\pi^2}+\dfrac{4L^2}{n^3\pi^3}\]
\[C_{1}=-\frac{2L^2\cos n \pi}{n \pi}\\C_{2}=\frac{4L^2\cos n \pi}{n \pi}-\frac{4L^2\cos n \pi}{n^3 \pi^3}+\frac{4L^2}{n^3\pi^3}\]
well you can cancel sin term in c_1
from then, I can construct U(x,t) , right?
\[C_{1}=-\frac{2L^2\cos n \pi}{n \pi}\\C_{2}=\frac{2L^2\cos n \pi}{n \pi}-\frac{4L^2\cos n \pi}{n^3 \pi^3}+\frac{4L^2}{n^3\pi^3}\]
yes, you can
It's too late here. I will redo and repost it, just the term I messed up (page 3 and 4) , Can you help me check mine tomorrow?
I'll try (:
One more question, that is the way I have to do, right? that 's fourier series, right?
hey, (: means???
I saw everybody use :) which stands for smiling face and :( for crying one, yours is (: I don't know
Anyway, thanks, thanks, thanks a ton.
smiling (:
is that the final answer? ok, let me try to get it.
where \(\alpha^2\)?
I used k
I forgot t, this is the correct one
my book uses different letter, don't be confused, use your letter..
aha, yours is k, got you. thanks again
@cinar I know you are offline, just post in case you came online, check mine, please. Compare to wolframalpha, my \(C_n\) is ok, but the last step when plug everything into the formula, I am not sure. My argument is whatever n is, sin npi =0 and cos npi =-1, therefore, the last result should be as mine.
when n is odd cosnpi=-1 n is even cosnpi=1 so you can replace cosnpi by (-1)^n I can't see your last sol.
\[ \Large C_{n}=\frac{-4(-1)^nL^2}{n^3 \pi ^3}+\frac{4L^2}{n^3 \pi^3}\\ \Large U(x,t)=\sum_{n=1}^{\infty}(\frac{-4(-1)^nL^2}{n^3\pi ^3}+\frac{4L^2}{n^3 \pi^3})e^{\frac{-n^2\pi^2\alpha^2}{L^2}t}sin{\frac{n\pi x}{L}}\]
Thank you, cinar
Join our real-time social learning platform and learn together with your friends!