Recovering function from its gradient

  • Thread starter Thread starter Nero26
  • Start date Start date
  • Tags Tags
    Function Gradient
Nero26
Messages
21
Reaction score
1
Hi all,
I've to recover a function ∅ such that F=∇∅,F=3x^2y i +(x^3+2yz)j+y^2k.
So, ∂∅/∂x=3x^2y, Integrating w.r.t. x
∅=x^3y+f(y,z) ,Assuming there may be function of y and z.----------1
∴∂∅/∂y=x^3+f'(y,z)
now, ∂∅/∂y=(x^3+2yz)=x^3+f'(y,z)
∴f'(y,z)=2yz
To find ∅ from 1 I've to find f(y,z).
My problem is how to get f(y,z) from f'(y,z),should I integrate it w.r.t. y or z?
Thanks for your help.
 
Last edited:
Physics news on Phys.org
Nero26 said:
Hi all,
I've to recover a function ∅ such that F=∇∅,F=3x^2y i +(x^3+2yz)j+y^2k.
So, ∂∅/∂x=3x^2y, Integrating w.r.t. x
∅=x^3y+f(y,z)

OK so far

,Assuming there may be function of y and z.----------1
∴∂∅/∂y=x^3+f'(y,z)
now, ∂∅/∂y=(x^3+2yz)=x^3+f'(y,z)
∴f'(y,z)=2yz


You shouldn't use ' for a partial derivative. What you have is ##\phi_y = x^3 +f_y(y,z)## which must equal ##x^3+2yz##.
To find ∅ from 1 I've to find f(y,z).
My problem is how to get f(y,z) from f'(y,z),should I integrate it w.r.t. y or z?
Thanks for your help.

And that illustrates perfectly why you should use the partial symbol instead of the '. So what you actually have to solve at this step is ##f_y(y,z) = 2yz##. Can you finish it From there?
 
Last edited:
LCKurtz said:
OK so far




You shouldn't use ' for a partial derivative. What you have is ##\phi_y = x^3 +f_y(y,z)## which must equal ##x^3+2yz##.


And that illustrates perfectly why you should use the partial symbol instead of the '. So what you actually have to solve at this step is ##f_y(y,z) = 2yz##. Can you finish it From there?

Thanks a lot for your help.
Now I got it, I've to integrate it w.r.t. y to get ##f_y(y,z)## to get f(y,z),Right?
Btw, I've got a new problem,F=i(x3z-2xyz)+j(xy-3x2yz)+k(yz2-zx) is a solenoidal vector field.I've to find a V such that F=∇XV
So,now it comes to find a vector field from its curl.How can I find it? In my book only from gradient was explained but not from curl :cry: So please help me out.
 
Don't know if there's a smarter way, but the obvious is to write out three PDEs, one for each component of F.
 
haruspex said:
Don't know if there's a smarter way, but the obvious is to write out three PDEs, one for each component of F.
Sorry,I exactly didn't get you.
I think PDE=partial differential equations.If so with respect to which variable x,y,or z or all of them?
Please say something more,
Thanks a lot for your help.
 
Nero26 said:
Sorry,I exactly didn't get you.
I think PDE=partial differential equations.
Yes. You know the definition of and of cross-product, so given the vector that results, F, you can write down three PDEs, one for each component of ∇XV. That gives you three equations involving ∂V/∂x etc.
 
There are two things I don't understand about this problem. First, when finding the nth root of a number, there should in theory be n solutions. However, the formula produces n+1 roots. Here is how. The first root is simply ##\left(r\right)^{\left(\frac{1}{n}\right)}##. Then you multiply this first root by n additional expressions given by the formula, as you go through k=0,1,...n-1. So you end up with n+1 roots, which cannot be correct. Let me illustrate what I mean. For this...
Back
Top