Ask your own question, for FREE!
Mathematics 7 Online
OpenStudy (anonymous):

Solve using power series. y" + 4y = 0

OpenStudy (anonymous):

No dude. would u help me with this then?

OpenStudy (anonymous):

yes I know the power series for e^x. i bet its 1 + x + x^2/2! + x^3/3! + .... but I don't know how to execute it. ;(

OpenStudy (nincompoop):

with a bullet in the head

OpenStudy (sirm3d):

let \[\Large y=\sum_{n=0}^{\infty} a_n x^n=a_0+a_1x+a_2x^2+\cdots\] for some constants \(a_n\) \[\Large y''=2(1)a_2+3(2)a_3 x+4(3)a_4 x^2+\cdots\] now for the equation \(\Large y''+4y=0\) \[\large 2(1)a_2+3(2)a_3x+4(3)a_4x^2+\cdots +4(a_0+a_1x+a_2x^2+\cdots) = 0\] for this equation to be true for all x, the coefficients should all equal zero, that is \[\Large{2a_2+4a_0=0\\3(2)a_3+4a_1=0\\4(3)a_4+4a_2=0\\\vdots}\]

OpenStudy (anonymous):

bullet to the stomach is more painful

OpenStudy (anonymous):

@kropot72 . thanks dear. So I have to find then how it arrives to the general solution.

OpenStudy (sirm3d):

\[\Large{2a_2+4a_0=0\Rightarrow a_2=-(4/2)a_0\\6a_3+4a_1=0\Rightarrow a_3=-(4/6)a_1\\12a_4+4a_2=0\Rightarrow a_4=-(4/12)a_2\\\vdots\\a_{n}=-(4/n(n-1))a_{n-2}\qquad,n=2,3,4,\dots}\]

OpenStudy (sirm3d):

@kropot72 , \[\Large y=Ae^{2x}+Be^{2x}\] is not the correct solution.

OpenStudy (anonymous):

use the power series \[\sum_{i=0}^\infty a_nx^n\] derivative of this will be step up one \[\frac{dy}{dx}=\sum_{i=1}^\infty na_nx^{n-1}\]

OpenStudy (anonymous):

so do that again and then what you want to do is get everything to the \[x^n\]

OpenStudy (anonymous):

once you do that you should get something of the such \[(n-1)(n-2)(n+1)+a_{n+1}+a_{n-2}=0\\]

OpenStudy (anonymous):

\[(n-1)(n-2)(n+1)+a_{n+1}+a_{n-2}=0\]

OpenStudy (anonymous):

isolate one of the a terms and then you'll get something of the suh \[a_{n+1}=\frac{a_{n-2}}{(n-1)(n-2)(n+1)}\]

OpenStudy (anonymous):

then you can plug in values for n to find the power series

OpenStudy (anonymous):

This question has a small trick, but i will let others finish, because they started before me, if you want i can put the final answer as i solved on a piece of paper...

OpenStudy (anonymous):

oh also you want the n=0 to be the smae for all in each in other words... you want to take the derivatives of that sum and then you'll get 2 different sums... those sums can then be added up to one sum

OpenStudy (anonymous):

All help is much appreciated. thanks alot. Still solving here, using all your ideas. Thanks ^_^

OpenStudy (anonymous):

Well as the guys mentioned and they have done a great job, you need to differentiate the sums....

OpenStudy (anonymous):

\[y = \sum_{n=0}^{\infty} a_nx^n\]

OpenStudy (anonymous):

\[y \prime = \sum_{n=1}^{\infty}n a_nx^{n-1}\]

OpenStudy (anonymous):

\[y'' =\sum_{n=2}^{\infty}n(n-1 )a_nx^{n-2}\]

OpenStudy (anonymous):

\[\sum_{n=2}^{\infty}n(n-1 )a_nx^{n-2} + 4\sum_{n=0}^{\infty} a_nx^n = 0\]

OpenStudy (anonymous):

this is done by plugging the series in y''+4y=0

OpenStudy (anonymous):

now the only trick in such questions is when you have different indexes in summations, and to solve such a problem we need to shift the sum in order to have the same starting point...

OpenStudy (anonymous):

as you can see we have n =0 in one summation and we have n =2 in the other one...

OpenStudy (anonymous):

so we need to shift n=2 to be n=0

OpenStudy (anonymous):

\[\sum_{n=0}^\infty (n+2)(n+1)a_{n+2}x^n+4\sum_{n=0}^\infty a_x x^n=0\]

OpenStudy (anonymous):

\[\sum_{n=0}^{\infty} (n+1)(n+2)a_{n+2}x^n\]

OpenStudy (anonymous):

can you see what happened? because we started from 2, but we need to start from 0, then we shift the sum by 2 units at every component in the series

OpenStudy (anonymous):

\[\sum_{n=0}^{\infty} (n+1)(n+2)a_{n+2}x^n+4\sum_{n=0}^{\infty} a_nx^n = 0\]

OpenStudy (anonymous):

am i disconnected my pc is going mad

OpenStudy (anonymous):

now what we can do, we factorize the x_n and rearrange the sums

OpenStudy (sirm3d):

\[\large \sum_{n=0}^{\infty}\left\{ (n+2)(n+1)a_{n+2} +4a_n\right\}x^n=0\] this means \[\large {(n+2)(n+1)a_{n+2}+4a_n=0\qquad n=0,1,2,\cdots\\a_{n+2}=-\frac{4}{(n+2)(n+1)}a_n}\]

OpenStudy (anonymous):

yes that is the right answer saved me time...

OpenStudy (anonymous):

but there is still more for this question... to come

OpenStudy (anonymous):

now from the power series, we will get some sort of relation, most likely similar to functions expanded using Taylor series...

OpenStudy (anonymous):

the philosophy behind such a system its a bridge between explicit ODE solution and Numerical Analysis for ODE

OpenStudy (anonymous):

So sirm3d, I solved down to \[a _{n+2}= \frac{ -4 a _{n} }{ (n+2)(n+1) }\] and got the series \[\sum_{n=0}^{\infty} \frac{ (-1)(2x)^{2n} }{ (2n)! }\] for the c0 terms and \[\sum_{n=0}^{\infty} \frac{ (-1)x ^{2n+1} }{ (2n+1)! }\] for the c1 terms. Is this correct?

Can't find your answer? Make a FREE account and ask your own questions, OR help others and earn volunteer hours!

Join our real-time social learning platform and learn together with your friends!
Can't find your answer? Make a FREE account and ask your own questions, OR help others and earn volunteer hours!

Join our real-time social learning platform and learn together with your friends!