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

Differential Equations Initial Value Problem \(ty\prime +2y=\sin(t)\) \(y(\frac{\pi}{2})=1,~t>0\)

terenzreignz (terenzreignz):

Let's see what we can work out... \[\Large y ' + \frac2t\cdot y = \frac{\sin(t)}t\]

terenzreignz (terenzreignz):

Doesn't seem to work out quite as intended...

OpenStudy (austinl):

Ok, so if we were to use the constant of integration it would be \(\mu(t)=2t\)?

terenzreignz (terenzreignz):

Oh yeah, it does LOL

terenzreignz (terenzreignz):

Okay, so we multiply everything by 2t... \[\Large 2ty'+ 4y = 2\sin(t)\]

terenzreignz (terenzreignz):

You can do it from here?

OpenStudy (austinl):

I was just learning this before the weekend and I haven't touched it in 3 days....

OpenStudy (austinl):

Do I integrate both sides? What do I do with the 4y? I am confusing myself.

terenzreignz (terenzreignz):

Okay... well, you'll notice something about the left side... it's just \[\LARGE \frac{d}{dt}(2ty)=2\sin(t)\]

terenzreignz (terenzreignz):

Using the product rule and implicitly differentiating 2ty with respect to t, you get 2ty' + 4y ^_^

terenzreignz (terenzreignz):

And now, just 'multiply both sides' by dt and you'll be on your merry way :D \[\Large d(2ty) = 2\sin(t)dt\]

terenzreignz (terenzreignz):

See where this is going? ^_^ \[\Large \int d(2ty) = \int 2\sin(t)dt\]

terenzreignz (terenzreignz):

Finish it ! :D

OpenStudy (austinl):

Indeed.

terenzreignz (terenzreignz):

So you should get \[\Large 2ty = -2\cos(t) + C\]

terenzreignz (terenzreignz):

Now we just need the value of that constant... that's not going to be too hard, right? :D

OpenStudy (austinl):

We would need to solve for y right?

terenzreignz (terenzreignz):

No... just C :) You know that \[\Large y\left(\frac \pi 2\right)= 1\] So, just have y = 1 and \[\large t = \frac \pi 2\] and solve for C

terenzreignz (terenzreignz):

And then after you find out exactly what value C takes, then you can do the finishing touches and solve for y so that it's a function. And you'd be done :D

OpenStudy (austinl):

\(C=\pi\)

terenzreignz (terenzreignz):

Okay :D So we have \[\Large 2ty = -2\cos(t) + \pi\]

OpenStudy (austinl):

\(y=\dfrac{-\cos(t)}{t}+\dfrac{\pi}{2t}\) Does this look even remotely correct?

terenzreignz (terenzreignz):

Rather, to make things look prettier... \[\Large 2ty = \pi - 2\cos(t)\]

terenzreignz (terenzreignz):

I think this is better : \[\Large y(t) = \frac{\pi - 2\cos(t)}{2t}\]

OpenStudy (austinl):

Okay, but a quick question. On a test, would I have been technically correct?

terenzreignz (terenzreignz):

With this: \[y=\dfrac{-\cos(t)}{t}+\dfrac{\pi}{2t}\]? Yes. Of course. I just like it when there's only one fraction bar :D

OpenStudy (austinl):

Okay, I just want to make sure! :D

terenzreignz (terenzreignz):

hang on a sec..

OpenStudy (austinl):

wanted*

OpenStudy (austinl):

Thank you very much for your help!!

terenzreignz (terenzreignz):

Yeah, just that Wolfie is giving me doubts...

OpenStudy (austinl):

My book lists this as it's answer. \(y=t^{-2}[~(\pi^2/4)-1-t \cos(t) +\sin(t)~]\)

terenzreignz (terenzreignz):

Yeah... my bad, wrong integrating factor.

terenzreignz (terenzreignz):

This is the correct integrating factor:\[\Large e^{\mu}\] where \[\Large \mu = \int \frac2tdt\]

terenzreignz (terenzreignz):

This just happens to be \[\Large \mu = 2\int \frac1tdt = 2\ln(t) = \ln (t^2)\]

OpenStudy (austinl):

\(\mu(t)=t^2\) I believe.

terenzreignz (terenzreignz):

Yup... you and your 2t -.- LOL Okay, so THAT's what we multiply to both sides... \[\Large t^2 y ' + 4ty = t\sin(t)\]

terenzreignz (terenzreignz):

---... I think I'm missing something...

OpenStudy (austinl):

I think I know.

terenzreignz (terenzreignz):

Oh yeah, \[\Large t^2y′+\color{red}2ty=t\sin(t)\]

OpenStudy (austinl):

\(t^2y\prime +2ty=2\sin(t)\)

OpenStudy (austinl):

Darn it :P

terenzreignz (terenzreignz):

So now, we have the left side as \[\Large \frac{d}{dt}(t^2y)= t\sin(t)\]

OpenStudy (austinl):

In all instances, will the middle term just disappear?

terenzreignz (terenzreignz):

Yes, that's kind of the point of the integrating factor...

OpenStudy (austinl):

Ok, sorry. I must seem so silly right now :P

terenzreignz (terenzreignz):

Say, this is general... \[\Large y' + p'(t)y= q(t)\]

terenzreignz (terenzreignz):

Now, the integrating factor would be \[\Large e^{p(t)}\]

terenzreignz (terenzreignz):

Multiplying everything through with this factor gives \[\Large e^{p(t)}y' + e^{p(t)}p'(t)y = q(t)e^{p(t)}\]

terenzreignz (terenzreignz):

And that thing on the left ALWAYS ends up as: \[\Large \frac{d}{dt}[e^{p(t)}y]= q(t)e^{p(t)}\]

terenzreignz (terenzreignz):

And hopefully, that term on the right (which is a function of t only) is easy to integrate :D

OpenStudy (austinl):

Okay, thanks very much! That makes much more sense. My professor wasn't entirely clear about that :)

terenzreignz (terenzreignz):

Shall we continue? :)

OpenStudy (austinl):

I have \(t^2y=\sin(t)-t\cos(t)+C\)

OpenStudy (austinl):

\(C=\dfrac{\pi^2}{4}-1\)

terenzreignz (terenzreignz):

And hopefully, you can see where this is going now? ^_^

OpenStudy (austinl):

Now for a final answer in a very rough form I have, \(y=\dfrac{\sin(t)}{t^2}-\dfrac{\cos(t)}{t}+\dfrac{\pi^2}{4t^2}-\dfrac{1}{t^2}\)

OpenStudy (austinl):

That can be cleaned up I am sure.

terenzreignz (terenzreignz):

But that is the correct answer, right?

OpenStudy (austinl):

I believe so.

terenzreignz (terenzreignz):

Then... let's get ice cream... to celebrate a job well-done :D

OpenStudy (austinl):

Woohoo!!!

OpenStudy (austinl):

You deserve many more medals than I can give :P

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!