Find the solution of the given Initial Value Problem.
\(y''+2y'+2y=h(t)\) \(y(0)=0\) \(y'(0)=1\) \[ h(t)=\{ \begin{array}{ccc} 1,&\pi\le t <2\pi ~~~~~~~~~~~0,&0\le t < \pi \end{array} \]
That is supposed to be a multi-line brace. I wasn't able to get it to display properly. It is supposed to be piecewise style...
hmm, heaviside seems to come to mind with piecewise stuff
what pat has you stymied?
*part ... different keyboard makes my typos even more abundant :/
I am looking at the book and it is showing parts where the h(t) is equal to functions of u sub some number of 't' and I am not sure where they are getting them.
any foureir stuff in the lesson by chance?
Nope, that isn't for another 4 chapters.
ut is usually something to do with finding a ... forget what they call it .... function.
I have the answer to the question, it is just a matter of how the heck do I get there :/
can you attach a picture of the problem and solution?
BVP, boundary value problem seems to be something to this
I am working on getting the pics on my PC now. Just a moment. Also, it is just an answer, not a worked out solution.
Here we are, problem 2 is my assigned problem.
Just so you know, I only need to do part (a) as well.
it looks like they are using u(t) as an arbitrary "constant" in the solution
I think so, but then in other examples, they have very specific subscripts which then carry over as part of how to solve the problem. I has me rather confused all around.
can you provide a very specific example? :)
http://tutorial.math.lamar.edu/Classes/DE/StepFunctions.aspx heaviside seems applicable
the subscripts just indicate where the pieces are redefined .... they represent an interval marking
Ok, do you understand this well enough that you could help me through this problem? I am reading through those notes and I can kinda see how they are using the subscripts, now I just need to know how to apply this to this problem.
http://tutorial.math.lamar.edu/Classes/DE/IVPWithStepFunction.aspx example 4 looks like a good run thru, without reading it, im assuming they convert the h(t) defined by the piecewise into its heaviside equivalent ..... have to read further to test that hypothesis :)
Ok, so going by their example I get something that looks like this, \(\large{h(t)=1+(-1)u_{2\pi}(t)}\)
Or would it be, \(\large{h(t)=1+(-1)u_{3\pi}(t)}\)
im having to do a review of the heaviside ... the u parts seems to identify a switch that turns on when its subscript is reached. h(t) = 0 , 0 <= t < pi f(t) = 0 , to start with .................................. h(t) = 1 , pi <= t < 2pi f(t) = 0 + (1)\(u_{pi}(t)\) ................................... h(t) = 0, 2pi <= t since the initial, and the first switch are on; we have a value of 1, we need a value of -1 when the switch is turned on to equal 0 again f(t) = 0 + (1)\(u_{pi}(t)\) - (1)\(u_{2pi}(t)\)
Okay, I think I am following...
In the example, it said something about a "shift", how would that be done with this problem?
in most cases, the piecewise function is not defined in terms of constant values, but as function. the shifting seems to be a way to stabalize functions .... if im reading it right
"most switches will not turn on and take constant values. Most switches will turn on and vary continually with the value of t."
your problem gives us constants so im assuming that its a simpler setup ...
I am reading this stuff, I am not seeing what they are doing to them to get this. It almost seems random, to me anyway.
the heaviside page explains alot of what they are brushing over since it came before this and is assumed to be prior experience :) http://tutorial.math.lamar.edu/Classes/DE/StepFunctions.aspx
\[L\{u_c(t)\}=\frac{e^{-cs}}{s}\]
\[L\{u_{pi}-u_{2pi}\}=\frac{e^{-pi~s}}{s}-\frac{e^{-2pi~s}}{s}\]
we also need to take the laplace of the equation \[L\{y''+2y'+2y\}\] you recall how to take the laplace of that?
\(\large{ s^2Y(s)-y(0)-y'(0)+2(sY(s)-y(0))+2(Y(s))=\frac{e^{-\pi s}}{s}-\frac{e^{-2\pi s}}{s}}\) This look right?
\[(s^2Y(s)-sy(0)-y'(0))+2(sY(s)-y(0))+2Y(s)\] filling in the initial conditions give us: \[s^2Y(s)-1+2sY(s)+2Y(s)\] factoring Y(s) \[(s^2+2s+2)Y(s)-1\] and solving for Y(s), gives us ... \[Y(s)=\frac{e^{-pi~s}-e^{-2pi~s}+s}{s(s^2+2s+2)}\]
yes, yes it does :)
i think i ran over a 2 someplace in mine
nah, y=0 so it didnt distribute worth anything ....
Okay.... I have.... \(\displaystyle Y(s)(s^2+2s+2)-1=h(t)\) I think this is similar to what you got... Where would I go from here?
add 1 and divide the s^2 poly
\(\displaystyle Y(s)=\frac{e^{-\pi x}}{s(s^2+2x+2)}-\frac{e^{-2\pi s}}{s(s^2+2x+2)}+\frac{1}{(s^2+2x+2)}\) This is where I am at....
good, now we can essentially ignore the e parts and work out the laplace inverses of:\[F(s)=\frac1{s(s^2+2s+2)}\]and\[G(s)=\frac1{s^2+2s+2}\]
On F(s) I believe we can even break that into two different parts as well, can't we?
a by parts decomp would be fine
notice that G(s) gives us the e^(-t) sin(t) part of the solution
Yep, I see that.
the partial decomp should provide the rest of the solution they have when we apply it to the upi and -u2pi inverses
without a lot of experience with heaviside and Laplace, i think that worked out pretty nicely :)
Yeah! Thanks very much, makes more sense to me now anyway :P
youre welcome :)
Join our real-time social learning platform and learn together with your friends!