How do I use this linearity, laplace
\[y''-2y'+2y=cost\]
@ganeshie8
\[s^2 \mathcal{L}(y)(s)+2s(y)+2y'(0)-\] yeah I have no clue how to even set this up
Initial conditions are y(0)=1, y'(0)=0
Oh wait that's a formula
\[y'(t)\leadsto sY(s)-y(0)\] \[y''(t)\leadsto s^2Y(s)-sy(0)-y'(0)\]
Let Y = L(y)(s) \[s^2Y-s f(0)-f'(0)+sY-f(0)=\frac{ s }{ s^2+1 }\]
and y
f and y two varuables are confusing... maybe stick to one variable
Then put the 2's
Let me rewrite it because that looks weird
\[y''-2y'+2y=\cos t\] taking laplace tranform both sides gives \[s^2Y(s)-sy(0)-y'(0)-2(sY(s)-y(0))+2Y(s)=\dfrac{s}{s^2+1}\]
Is this always true \[f'' = s^nY(s)-sy(0)-y'(0)\] and then same for f'
s^2
that is the formula which you must derive upfront
you must have formulas for the laplace transforms of first and second derivatives before starting to solve differential equations using laplace transforms
Oh ok, I was just confused by the y'' and y' I didn't know what to put, but I see it's on the table
These are the formulas : \[y'(t)\leadsto sY(s)-y(0)\] \[y''(t)\leadsto s^2Y(s)-sy(0)-y'(0)\] you can derive them using the definition of laplace transform and IBP
may be lets derive the laplace transform of first derivative
\[f^{n}(t) = s^nF(s)-s^{n-1}f(0)-...-f^{n-1}(0)\]
\[\mathcal{L}(f'(t))=?\]
plug that in the definition of laplace transform (integral) and play with it a bit
Into the integral right
The one we were just working on earlier
yes
Oh I see it would give you the fundamental theorem of calc
watch between 20th minute and 30th minute
sL(f(t))-f(0)
you love this guy
So good, dv = f'(t) in this case
a variable constant XD
\[\frac{ s^3+2s^2+2s+2}{ (s^2+1)(s^2-2s+2) }\] wow
sign mistake nooo
parth help
\[s^2 \mathcal{L}(y)(s)+2s(y)+2y'(0)\]it is a quadratic expression
get out
lol
\[\frac{ s^3-2s^2+2s-2}{ (s^2+1)(s^2-2s+2) }\]
what do you want to do with this
there we go
now we do partial fractions decomposition
\[\frac{ s^3-2s^2+2s-2}{ (s^2+1)(s^2-2s+2) } = \frac{ As+B }{ (s^2+1) }+\frac{ Cs+D }{ (s^2-2s+2) }\]
#wolframtime
lol he does those terrible jokes when he feels good and he feels good almost all the time
That was a really good video though
im talking about prof matt..
I konow
hey that was a beautiful gif where did it go
|dw:1448443636882:dw|
Ok I think I should be good, just needed to know how to set it up, so couldn't we do this with undetermined coeff, variation of parameter
Why exactly is laplace good that all the other methods included what I mentioned and series can't do
all the methods you have included are also good
they do the job pretty well too
engineers like laplace transforms because they always know initial conditions and laplace transforms give the answer one shot
and it's more plug and chug
compare that with the pain of finding complementary and parituclar solutions which are required with all other methods thats you have listed earlier
trueeee
undetermined coeff is probably the worst
Also the method itself is just awesome. It changes the problem of solving "differential equation" to a problem of solving "algebraic equation"
because you have to guess
haha yeah!
\[y''-2y'+2y=\cos t\] taking laplace tranform both sides gives \[s^2Y(s)-sy(0)-y'(0)-2(sY(s)-y(0))+2Y(s)=\dfrac{s}{s^2+1}\] do take time time to appreciate how easy it is to solve \(Y(s)\) in above equation
ofcourse finding inverse laplace transform is again a pain
Factor and isolate, then use inverse
Yeah
IS this right, \[\frac{ s^3-2s^2+2s-2}{ (s^2+1)(s^2-2s+2) } = \frac{ As+B }{ (s^2+1) }+\frac{ Cs+D }{ (s^2-2s+2) }\] or do I have to go back into my calc text
s^2-2s+2 can be done by completing the square I guess ill have to do that for the final solution but thats easy
just double check numerator once i don't have pen and paper wid me as of now..
I think it's good, let me see if i get the final answer as they have in the back
im getting s^3-2s^2+s-2 in the numerator
\[Y(s)(s^2-2s+2)-s+2 = \frac{ s }{ s^2+1 }\] \[Y(s)(s^2-2s+2) = \frac{ s }{ s^2+1 }+s-2\] \[Y(s)(s^2-2s+2)= \frac{ s^3-2s^2+2s-2 }{ s^2+1 }\] mhm
ohk my mistake... pls keep going :)
Is there no formula for \[\frac{ 2 }{ s^2+1 }\]
I just used wolfram to find the coeff |dw:1448444960745:dw|
that looks good
Oh good!
Wolfram, our Lord and Savior
finding inverses of these two must be easy |dw:1448445070422:dw|
Join our real-time social learning platform and learn together with your friends!