# solving initial value problem!

Show 40 post(s) from this thread on one page
Page 1 of 2 12 Last
• Nov 18th 2012, 09:38 PM
tjsdndnjs
solving initial value problem!
I have

xy' - 3y = x^4(e^x + cos x) - 2x^2 ; y(pi) = e^pi + (2/pi).

Could someone solve this out for me?

Thank you.
• Nov 18th 2012, 09:48 PM
MarkFL
Re: solving initial value problem!
I'm not going to post a full solution, but I will be glad to help you get one. Can you think of a way to write the ODE as a linear equation?
• Nov 19th 2012, 02:12 AM
hollywood
Re: solving initial value problem!
By linear, he means an equation of the form $y'+p(x)y = q(x)$. But that's easy:

$y' - \frac{3}{x}y = x^3(e^x + \cos {x}) - 2x$

Now, how do you solve a first-order linear equation?

- Hollywood
• Nov 20th 2012, 06:34 PM
tjsdndnjs
Re: solving initial value problem!
So i set p(x)= -3/x

integrate p(x)dx = - integrate 3/x dx = - ln x^3

then I did e^(-ln x^3) = 1/x^(3) which is the integrating factor... what do I do next??
• Nov 20th 2012, 06:56 PM
MarkFL
Re: solving initial value problem!
Multiply the equation in standard linear form by this integrating factor. You should then find the left-hand side can be expressed as the derivative of a product.

The equation in standard linear form:

$\frac{dy}{dx}-\frac{3}{x}y=x^3(e^x+\cos(x))-2x$

Now, multiply through by $\mu(x)=\frac{1}{x^3}$

$\frac{1}{x^3}\cdot\frac{dy}{dx}-\frac{3}{x^4}y=e^x+\cos(x)-\frac{2}{x^2}$

Rewrite the left side as the derivative of a product:

$\frac{d}{dx}\left(\frac{1}{x^3}\cdot y \right)=e^x+\cos(x)-\frac{2}{x^2}$

Now, integrate with respect to $x$.
• Nov 20th 2012, 07:02 PM
tjsdndnjs
Re: solving initial value problem!
Hello, I have a question how did 1/x^3 x dy/dx - 3y/x^4 became d/dx(1/x^3 x y) ??
• Nov 20th 2012, 07:18 PM
MarkFL
Re: solving initial value problem!
If the integrating factor is calculated correctly, you should always get the left side to be the derivative of the product of the integrating factor and the dependent variable.

Observe that:

$\frac{d}{dx}\left(\frac{1}{x^3}\cdot y \right)=\frac{1}{x^3}\cdot\frac{dy}{dx}-\frac{3}{x^4}y$

Your textbook should give you a development of this method which will demonstrate why this always works.

On a side note, I highly recommend that you do not use the character "x" to denote multiplication, as it can be confused with the variable $x$.
• Nov 20th 2012, 07:45 PM
tjsdndnjs
Re: solving initial value problem!
I integrated the right side respect to x then i got

y/x^3 = e^x + cos x - 2/x + c

Then I multiplied x^3 to both side getting

y= x^3 e^x + x^3 sin x + 2/x^2 + c/x^3

Am I right so far? because I have this equation y(pi) = e^pi + (2/pi)

But I tried and it does not seem to work
• Nov 20th 2012, 08:04 PM
MarkFL
Re: solving initial value problem!
You have integrated the trigonometric term on the right incorrectly (wait, I see you changed it in the second line), so not counting you first line, what you have done incorrectly is you have integrated the rational term incorrectly and divided the parameter (constant of integration) by x^3 rather than multiplied. So, what you should have is the general solution:

$y(x)=x^3\left(e^x+\sin(x)+\frac{2}{x}+c \right)$

Now, use the given initial condition to determine $c$.
• Nov 20th 2012, 08:11 PM
tjsdndnjs
Re: solving initial value problem!
Deos the constant C becomes Cx^3? after multiplying?
• Nov 20th 2012, 08:23 PM
MarkFL
Re: solving initial value problem!
Yes, if you choose to distribute, then everything withing the parentheses gets a factor of x^3.
• Nov 20th 2012, 10:46 PM
tjsdndnjs
Re: solving initial value problem!
Hello so I got this equation

e^pi + 2/pi = pi^3 e^pi + 2pi^2 +Cpi^3 from y(pi) = e^pi + 2/pi

I tried moving everything to left leaving Cpi^3 on the right. But I find that the equation gets very messy? Is it supposed to be like that?
• Nov 20th 2012, 10:56 PM
MarkFL
Re: solving initial value problem!
Yes, the value I found for C is not a simple number. Post what you find for C and I will tell you if our results agree.
• Nov 20th 2012, 10:59 PM
tjsdndnjs
Re: solving initial value problem!
I have got (pi e^pi - 2 - pi^4 e^pi - 2pi^3)/pi^4
• Nov 20th 2012, 11:50 PM
MarkFL
Re: solving initial value problem!
That is similar to what I have, not not the same. Could you post your derivation?
Show 40 post(s) from this thread on one page
Page 1 of 2 12 Last