Differential Equation PLEASE HELP!
\[\frac{ (y-xy') }{ (x+yy' }\]
=2
\[\frac{ (y-xy') }{ x+yy') }=2\]
That's not ODE, is it?
@rvc?
@IrishBoy123 ?
rearrange it so that it reads y' = ...whatever you get then we'll call the result y' = f(x, y) if then f(tx, ty ) = f(x,y), then it is a homo and you are good to go.
I got \[y'=\frac{ (2x+y) }{ (y-2x) }\]
you mean to write y' as f(2x,y)?
is that not upside down?! or have i misread the original equation.
Was I wrong even with that?
\[\frac{ y - xy' }{ x + y y' } = 2\] y - xy' = 2(x + yy') y - xy ' = 2x + 2yy' y'(2y +x) = y - 2x
ok, but what m I doing from here?
am
y' = f(x,y) = (y - 2x)/(2y +x) yes?
so the homo test is: does f(kx,ky) = f(x,y) for any number k. can you establish that that is true, because it means we can nail it using a homo sub.
so k is 2 right?
no, k is any number. replace x by kx and y by ky in f(x,y) to get f(kx, ky). and find out if they are still equal.
yes
then homo sub right?
yes
so the fraction should be replaced with U ?
is that what you mean?
no, now you make the sub y = v x, ie replace all y's with vx. then, because y = v x, you know that y' = xv' + v , and so you can make that sub too. that is now separable in v and x, though there is some algebra to be done, moving the v's to the LHS and the x to the RHS. bit messy but totally do-able..
I hope so, I'm doing it right now
@IrishBoy123 it's not working...
@phi?
once you get to \[ y' =\frac{ (y - 2x)}{(2y +x)} \] you do what you did on previous problems y = ux y' = u + x u' how far do you get ?
yes
but it didn't work from there and I don't know why...
how far did you get ?
a mess, one stage after that
I'll try again one sec
\[u'x+u=\frac{ (ux-2x) }{ (2ux+x) }\]
you can factor out x from top and bottom on the right, and cancel. write u' as du/dx and separate variables. what do you get ?
\[\frac{ du }{ x }+u=\frac{ (u-2x) }{(2u+1)}\]
that does not look correct
I did common denominator, but it didn't help much tbh
first, just simplify the right side: \[ \frac{ (ux-2x) }{ (2ux+x) } \]
ups
?
I got this \[\frac{ (2u+1) }{ (u-2) }du=\frac{ dx }{ x }\]
it seems better now ehe?
closer. But you have to be a lot more careful with your algebra \[ u'x+u=\frac{ (ux-2x) }{ (2ux+x) } \\ \frac{du}{dx} x+u=\frac{ (u-2) }{ (2u+1) } \\ \frac{du}{dx} x=\frac{ (u-2) }{ (2u+1) } -u \]
use a common denominator of 2u+1 on the right side
so what from here?
can you combine terms on the right side ?
yes \[\frac{ du }{ dx}=\frac{ -2-2u^2 }{ 2u+1}\]
ok, but remember there is a factor of x on the left side.
\[\frac{ (2u+1) }{ -2-2u^2 }du=dx\]
you started with \[ \frac{du}{dx} x=\frac{ (u-2) }{ (2u+1) } -u \] You lost the "x" ??
it's such a mess
messy or not, you need to keep the x
Should I start from the beginning?
This is ok \[ \frac{ du }{ dx}x=\frac{ -2-2u^2 }{ 2u+1} \] I would "flip" both sides \[ \frac{ dx }{ x}\frac{1}{du}=\frac{ 2u+1} { -2-2u^2} \]
what for?
because it (should be) clear that we want dx/ x on one side (surely not x/dx , right ?) and to get dx/x , the easy way is "flip" both sides then move the du to the other side.
in other words, we know how to integrate dx/x and not x/dx
ok
\[\frac{ (2u+1) }{-2-2u^2 }du=\frac{ dx }{ ? }\]
Affirmative?
why the question mark where the x should be ?
there should be an x
\[ \frac{ (2u+1) }{-2-2u^2 }du=\frac{ dx }{ x } \] we can integrate both sides, but obviously, the right side is easier just to "clean up" the left side I would factor -2 from the denominator: \[ - \frac{1}{2}\frac{ (2u+1) }{u^2+1 }du=\frac{ dx }{ x } \]
we have, so far \[ - \frac{1}{2} \int \frac{2u\ du}{u^2+1} - \frac{1}{2} \int \frac{du}{u^2+1} = \int \frac{dx}{x}\]
arctg?
twice?
the first is dw/w with w= u^2+1
so ln(u^2+1)*-0.5
for the first term
and then -0.5arctgu
yes
but the answer is \[\ln(x^2+y^2)+\arctan(y/x)=c\]
once you get the integrated expression, you can twist it into many different forms with your algebraic skills
wow thanks allot! @phi you always help me with so much patience
thank you!
It does not look particularly easy to get your book answer. Give it a try.
I shall
@phi?
I still don't see how do i get the answer....
@rvc is this as the answer?
ln(x2+y2)+arctan(y/x)=c that's the answer
but we got smth quite different
First, write down what you got after integrating, including the constant of integration
I got \[\frac{ -1 }{ 2 }\ln(\frac{ y }{ x })^2-0.5\arctan(\frac{ y }{ x })=lnx+c\]
is that correct?
now look at what you are trying to get, namely ln(x2+y2)+arctan(y/x)=c it's clear we want all the terms except the constant on one side so move the ln x to the left side
ok
but still
it's not the same
btw, how did you get that ln term it started as -1/2 ln (u^2 +1) with u= y/x
U=y/X
no?
yes, u =y/x . what is u^2 + 1?
it's the integral
\[\int\limits \frac{ dx }{x}=lnx\]
I mean, what is the expression u^2 + 1 after subbing in y/x for u
Join our real-time social learning platform and learn together with your friends!