Differential equations First order lnear
n is constant
looks like bernouli
you think so? Hmm
any idea how to solve this? =)
n is a constant means it could be 2 or 3 or some other number... it doesn't change as x or y change..
its telling us not to bother about "n"
Ohh. Thank you so much.
can you finish off the rest ?
Wait can i just multiply the whole equation with 1/nx?
which equation ?
so that i'll arrive with dy/dx + 2y/nx = xy^n=1 / nx ?
the thing i posted. hehe
it'll look like the gen formula for first order linear
\[\dfrac{y'}{y^{n+1}}+ \dfrac{2}{nx}\frac{1}{y^n} = \frac{1}{n} \] this is our equation, right ?
yes yes
but the orig eq
what i mean is that can i just multiply the orig equation with 1/nx?
yes, what good is that ?
\[nxy'+ 2y = xy^{n+1}\] multiplying throguh out by 1/nx you get \[y'+ \dfrac{2}{nx}y = \frac{1}{n}y^{n+1}\]
this is NOT a linear equation because the right hand side has \(y\) terms
for a linear wquation, the right hand side needs to be function of \(x\) alone, right ?
yes =) thank you
so ill integrate now the 2/nx right?
so that ill have the integrating factor
it wont work
Again, for integrating factor method to work, the right hand side needs to be a function of \(x\) alone
\[y'+ \dfrac{2}{nx}y = \frac{1}{n}\color{Red}{y^{n+1}}\] thats the bad part
yes youre right
A linear equation needs to be of below form : \[y' + p(x) y = q(x)\]
NO \(y\) terms on right hand side ok
so you can't use integrating factor method for your equation yet..
try it, it simply wont work
thats where bernouli substitution is helpful
what would be the px?
\[y'+ \dfrac{2}{nx}y = \frac{1}{n}\color{Red}{y^{n+1}}\] Again, this is NOT a lienar equation yet
integrating factor stuff wont work
yes hehe sorry
v= y ^ 1-n, right? in bern eq?
but we can make it linear, lets start by dividing \(y^{n+1}\) both sides
\[y'+ \dfrac{2}{nx}y = \frac{1}{n}\color{Red}{y^{n+1}} \] dividing through out by \(y^{n+1}\) gives \[\dfrac{y'}{y^{n+1}}+ \dfrac{2}{nx}\dfrac{1}{y^n} = \frac{1}{n}\]
Now substitute \(\large v = \frac{1}{y^n}\) differentiating both sides with respect to x gives \(\large v' = -\dfrac{n}{y^{n+1}} y' \implies -\dfrac{v'}{n} = \dfrac{y'}{y^{n+1}}\) the equation becomes \[-\dfrac{v'}{n} + \dfrac{2}{nx}v = \frac{1}{n}\] which is same as \[v'+ \dfrac{-2}{x}v = -1\]
Now thats a linear equation and you can use ur favorite integrating factor method :)
hahaha thank you so much. px would be -2/x right?
Yep!
is it okay to use the formula of bernoulli?
what formula ?
it should be okay to use any formula if it makes sense to you.. but i prefer this dividing and substitution process more as it lets me see whats going on..
hi c;
do you mean you got \(\color{Red}{v}\)x^-2 = 1/x +c
yes
plug back the value of \(v\) : \[v = \frac{1}{y^n}\]
the final solution will be \[\frac{1}{y^n}x^{-2} = \frac{1}{x}+ c\]
you may make it look better by multiplying x^2 through out
omg got it lalala haha thank you so much thank you
=) @ShadowLegendX
the final solution will be \[\frac{1}{y^n} =x+ cx^2\]
@ganeshie8 is da honey of all them bunnies He knows dat stuff like all his rabbit holes
Awesome
Join our real-time social learning platform and learn together with your friends!