Integration of a composite function

Tensel
Messages
6
Reaction score
0
y=f(x),K=F(y),
and dF(y)/dy=y^2/y', (1)
then
dF(x)=y^2*dx;
so, F(x)=int(y^2*dx)=int((f(x)^2)*dx);
then we obtain,
F(x)=(f(x))^3/(3*f'(x))+C;
substitution of y=f(x) into F(x), we get, F(y)=y^3/(3*y')+C; (2)

using the result above(eq.(2)), the dF(y)/dy can be computed as follow:

dF(y)/dy=y^2/y' - (y^3*y'')/(3*(y')^3) (3)

Probelm: Eq.(3) is not equal to the eq.(1).

I don't know which step is not correct.
 
Physics news on Phys.org
I have trouble figuring out which part of your calculation is correct!
 
Tensel said:
y=f(x),K=F(y),
What reason do you have for introducing "K" when it is not used again?

and dF(y)/dy=y^2/y', (1)
I don't believe this is possible. Do you mean y' to be the dervative of y with respect to x?
The derivative of F with respect to y has nothing to do with y's dependence of x.

If you meant dF/dx, then dF/dx= (dF/dy)(dy/dx). If you meant that F(y)= y^3/3, then dF/dx= y^2 times y', not divided by it.

then
dF(x)=y^2*dx;
and this doesn't follow from what you wrote before. I don't know where you got the y^2 but, again, if you meant that F(y)= y^3/3 then dF= y^3 y'(x)dx.

so, F(x)=int(y^2*dx)=int((f(x)^2)*dx);
then we obtain,
F(x)=(f(x))^3/(3*f'(x))+C;
substitution of y=f(x) into F(x), we get, F(y)=y^3/(3*y')+C; (2)

using the result above(eq.(2)), the dF(y)/dy can be computed as follow:

dF(y)/dy=y^2/y' - (y^3*y'')/(3*(y')^3) (3)

Probelm: Eq.(3) is not equal to the eq.(1).

I don't know which step is not correct.
 
Last edited by a moderator:
HallsofIvy said:
What reason do you have for introducing "K" when it is not used again?.
Yes, you are correct that K is not used, we can get rid of it.


I don't believe this is possible. Do you mean y' to be the dervative of y with respect to x?.
Yes, y' means the derivative of y with respect ot x.

The derivative of F with respect to y has nothing to do with y's dependence of x..
So, you meant that we should consider y' as constant when computing the derivative of F with respect to y? I need more information to understand this.

If you meant dF/dx, then dF/dx= (dF/dy)(dy/dx). If you meant that F(y)= y^3/3, then dF/dx= y^2 times y', not divided by it.

and this doesn't follow from what you wrote before. I don't know where you got the y^2 but, again, if you meant that F(y)= y^3/3 then dF= y^3 y'(x)dx.
At the beginning, we have dF(y)/dy=y^2/y', then dF(y)/dy=y^2/(dy/dx),
eliminate the dy on the two side of the equation, we obtain dF(y)=y^2*dx,
substitution of y=f(x) into it, dF(f(x))=dF(x)=y^2*dx
 
Tensel said:
So, you meant that we should consider y' as constant when computing the derivative of F with respect to y? I need more information to understand this.

Suppose that ##F(y) = y^3 / 3## and ##y## is a function of x. Then
##\frac{dF}{dy} = y^2##
##\frac{dF}{dx} = y^2 \frac{dy}{dx}##

At the beginning, we have dF(y)/dy=y^2/y',

How? Where did this come from?

then dF(y)/dy=y^2/(dy/dx), eliminate the dy on the two side of the equation, we obtain dF(y)=y^2*dx,

You are using ##\frac{1}{dy/dx} = \frac{dx}{dy}## which requires y to be an invertible function.
 
pwsnafu, thank you for your reply.
dF(y)/dy=y^2/y'-----this equation is what I want to construct, it can be considered as a known condition, I want to get F(y).
 
Last edited:
I find the problem.
given that, \frac{dF(y)}{dy}=\frac{y^{2}}{y'}
then,
\frac{dF(x)}{dx}=\frac{dF(y)}{dy}.\frac{dy}{dx}=\frac{y^{2}}{y'}.y'=y^{2},
so,
F(x)=\int y^{2}.dx=\int f^{2}(x).dx

I made a mistake in the integration of F(x)
i.e. F(x)≠\frac{f^{3}(x)}{3.f'(x)}+C

But, what should be the answer for F(x) from which I can obtain F(y)? should the specific expression for f(x) be given?
 
Last edited:
Back
Top