Vector analysis problem about a gradient

Click For Summary

Homework Help Overview

The discussion revolves around a vector calculus problem involving the gradient of a scalar function, specifically examining the relationship between a function defined in terms of scaled variables and its gradient. The original poster expresses confusion regarding the problem statement and seeks guidance on how to approach it.

Discussion Character

  • Exploratory, Conceptual clarification, Mathematical reasoning, Assumption checking

Approaches and Questions Raised

  • Participants discuss the correct formulation of the function and its implications. There are attempts to differentiate the function with respect to a scaling factor and questions about the relevance of the gradient in the context of the problem. Some participants suggest using the chain rule for differentiation.

Discussion Status

Participants are exploring various differentiation techniques and clarifying the relationship between the function and its gradient. There is a mix of understanding and confusion, with some participants providing insights while others seek further clarification. The discussion is active, with multiple interpretations being considered.

Contextual Notes

There is mention of specific conditions under which the function is defined, including the domain of the variables involved. Participants question the assumptions made in the problem statement and the implications of differentiating with respect to different variables.

patric44
Messages
308
Reaction score
40
Homework Statement
if Φ=Φ(λx,λy,λz) = λ^n(x,y,z)
proof that r.grad(Φ) = nΦ
Relevant Equations
Φ=Φ(λx,λy,λz) = λ^n(x,y,z)
r.grad(Φ) = nΦ
hi guys i saw this problem in my collage textbook on vector calculus , i don't know if the statement is wrong because it don't make sense to me
so if anyone can help on getting a hint where to start i will appreciate it , basically it says :
$$ \phi =\phi(\lambda x,\lambda y,\lambda z)=\lambda^{n}(x,y,z) $$
prove that
$$\vec{r} . \vec{\nabla}\phi=n\phi$$
 
Physics news on Phys.org
The correct statement is $$f(\lambda x)=\lambda^s f(x),\quad \forall \lambda>0,\quad \forall x\in D\subset \mathbb{R}^m.$$
Here the domain ##D## is such that ##x\in\ D\Longrightarrow \lambda x\in D##.Differentiate this equation in ##\lambda## and then put ##\lambda=1##:
$$\Big(x,\frac{\partial f}{\partial x}\Big)=sf(x).$$
 
  • Informative
Likes   Reactions: etotheipi
wrobel said:
The correct statement is $$f(\lambda x)=\lambda^s f(x),\quad \forall \lambda>0,\quad \forall x\in D\subset \mathbb{R}^m.$$
Here the domain ##D## is such that ##x\in\ D\Longrightarrow \lambda x\in D##.Differentiate this equation in ##\lambda## and then put ##\lambda=1##:
$$\Big(x,\frac{\partial f}{\partial x}\Big)=sf(x).$$
why should i differentiate with respect to ##\lambda## the gradient of ##\phi## is with respect to the ##(x,y,z)## ?
and what about ##\vec{r}.\vec{\nabla}\phi##
 
patric44 said:
why should i differentiate with respect to ##\lambda## the gradient of ##\phi## is with respect to the ##(x,y,z)## ?
and what about ##\vec{r}.\vec{\nabla}\phi##
You can let ##g(\lambda) = f(\lambda x, \lambda y, \lambda z) = \lambda^n f(x, y, z)## and differentiate that with respect to ##\lambda##.
 
  • Informative
Likes   Reactions: etotheipi
PeroK said:
You can let ##g(\lambda) = f(\lambda x, \lambda y, \lambda z) = \lambda^n f(x, y, z)## and differentiate that with respect to ##\lambda##.
ok :
let ##g(\lambda) = f(\lambda x ,\lambda y,\lambda z)=\lambda^{n}f(x,y,z)## then by differentiating :
$$\frac{dg(\lambda)}{d\lambda}=n\lambda^{n-1}f(x,y,z)$$
and the gradient part :
$$\vec{r}.\vec{\nabla}f(\lambda x ,\lambda y,\lambda z) = \lambda^{n}(x\frac{df}{dx}+y\frac{df}{dy}+z\frac{df}{dz})$$
i don't get how this is related to my question!
 
patric44 said:
ok :
let ##g(\lambda) = f(\lambda x ,\lambda y,\lambda z)=\lambda^{n}f(x,y,z)## then by differentiating :
$$\frac{dg(\lambda)}{d\lambda}=n\lambda^{n-1}f(x,y,z)$$
and the gradient part :
$$\vec{r}.\vec{\nabla}f(\lambda x ,\lambda y,\lambda z) = \lambda^{n}(x\frac{df}{dx}+y\frac{df}{dy}+z\frac{df}{dz})$$
i don't get how this is related to my question!
There's another way you can differentiate ##g(\lambda)##.
 
PeroK said:
There's another way you can differentiate ##g(\lambda)##.
a little more help , you mean using something like the chain rule ?
 
patric44 said:
a little more help , you mean using something like the chain rule ?
Yes, because the chain rule invokes the gradient.
 
PeroK said:
Yes, because the chain rule invokes the gradient.
:smile: thanks , i guess i got it now :
if i let ##u=\lambda x , v = \lambda y , w = \lambda z , ##then
$$\frac{dg}{d\lambda} = \frac{\partial\phi}{\partial u}\frac{\partial u}{\partial x}+\frac{\partial\phi}{\partial v}\frac{\partial v}{\partial y}+\frac{\partial\phi}{\partial w}\frac{\partial w}{\partial z}$$
then
$$\frac{dg}{d\lambda} = \frac{\partial\phi}{\partial \lambda x} x +\frac{\partial\phi}{\partial \lambda y}y+\frac{\partial\phi}{\partial \lambda z}z=n\lambda^{n-1}\phi(x,y,z)$$
if i let ##\lambda = 1## :
$$ \frac{\partial\phi}{\partial x} x +\frac{\partial\phi}{\partial y}y+\frac{\partial\phi}{\partial z}z=n\phi(x,y,z)=\vec{r}.\vec{\nabla}\phi$$
thanks guys so much .
 
  • #10
patric44 said:
:smile: thanks , i guess i got it now :
if i let ##u=\lambda x , v = \lambda y , w = \lambda z , ##then
$$\frac{dg}{d\lambda} = \frac{\partial\phi}{\partial u}\frac{\partial u}{\partial x}+\frac{\partial\phi}{\partial v}\frac{\partial v}{\partial y}+\frac{\partial\phi}{\partial w}\frac{\partial w}{\partial z}$$
then
$$\frac{dg}{d\lambda} = \frac{\partial\phi}{\partial \lambda x} x +\frac{\partial\phi}{\partial \lambda y}y+\frac{\partial\phi}{\partial \lambda z}z=n\lambda^{n-1}\phi(x,y,z)$$
if i let ##\lambda = 1## :
$$ \frac{\partial\phi}{\partial x} x +\frac{\partial\phi}{\partial y}y+\frac{\partial\phi}{\partial z}z=n\phi(x,y,z)=\vec{r}.\vec{\nabla}\phi$$
thanks guys so much .
This is not quite right. First, you should have:
$$\frac{dg}{d\lambda} = \frac{\partial\phi}{\partial u}\frac{\partial u}{\partial \lambda}+\frac{\partial\phi}{\partial v}\frac{\partial v}{\partial \lambda}+\frac{\partial\phi}{\partial w}\frac{\partial w}{\partial \lambda }$$
And then this becomes:
$$\frac{dg}{d\lambda} = \frac{\partial\phi}{\partial u}x +\frac{\partial\phi}{\partial v} y +\frac{\partial\phi}{\partial w} z = \vec{\nabla}\phi(u, v, w) \cdot \vec r = \vec{\nabla}\phi(\lambda x, \lambda y, \lambda z) \cdot \vec r$$
 
  • Like
Likes   Reactions: etotheipi
  • #11
PeroK said:
This is not quite right. First, you should have:
$$\frac{dg}{d\lambda} = \frac{\partial\phi}{\partial u}\frac{\partial u}{\partial \lambda}+\frac{\partial\phi}{\partial v}\frac{\partial v}{\partial \lambda}+\frac{\partial\phi}{\partial w}\frac{\partial w}{\partial \lambda }$$
And then this becomes:
$$\frac{dg}{d\lambda} = \frac{\partial\phi}{\partial u}x +\frac{\partial\phi}{\partial v} y +\frac{\partial\phi}{\partial w} z = \vec{\nabla}\phi(u, v, w) \cdot \vec r = \vec{\nabla}\phi(\lambda x, \lambda y, \lambda z) \cdot \vec r$$
oh my bad , but the book stated it explicitly as ##\vec{r}.\vec{\nabla\phi}## so i thought the book was right , does it matter since i can change the order of the product of the chain rule terms , and get the other reversed expression , or i cannot do that ?
 
  • #12
You still have to set ##\lambda = 1##. That sorts everything out.
 
  • Like
Likes   Reactions: patric44
  • #13
thanks so much :smile:
 
  • Like
Likes   Reactions: PeroK
  • #14
the converse theorem also holds
 
Last edited:

Similar threads

  • · Replies 9 ·
Replies
9
Views
2K
  • · Replies 8 ·
Replies
8
Views
2K
Replies
3
Views
2K
  • · Replies 6 ·
Replies
6
Views
3K
  • · Replies 1 ·
Replies
1
Views
2K
Replies
6
Views
2K
Replies
7
Views
2K
  • · Replies 7 ·
Replies
7
Views
2K
  • · Replies 5 ·
Replies
5
Views
2K
Replies
9
Views
2K