1

According to this wikipedia, in section Euler–Lagrange equation, in the first example (shortest path between two points), it says

$${\displaystyle {\frac {\partial L}{\partial f}}-{\frac {d}{dx}}{\frac {\partial L}{\partial f'}}=0}$$

with

$${\displaystyle L={\sqrt {1+[f'(x)]^{2}}}\,.}$$

Since f does not appear explicitly in L , the first term in the Euler–Lagrange equation vanishes for all f (x) and thus,

$${\displaystyle {\frac {d}{dx}}{\frac {\partial L}{\partial f'}}=0\,.}$$

..

I am confused. Are f and f' independent so that the first term in the Euler–Lagrange equation vanishes for all f (x)?

In short, why is $${\displaystyle {\frac {\partial L}{\partial f}}=0}$$ where $${\displaystyle L={\sqrt {1+[f'(x)]^{2}}}\,.}$$ ?

As $${\displaystyle {\frac {\partial L}{\partial f}}={\frac {\partial f'}{\partial f}}×{\frac {\partial L}{\partial f'}}}$$, does this mean that $${\displaystyle {\frac {\partial f}{\partial f'}}=0}$$ ?

Thank you in advance for making it clear for me.

  • You don't need to think of them as variables like that. $L$ is a scalar function of two scalar variables and its partial derivatives are taken with respect to those variables. The identity of the variables only matters after you have already assembled the E-L equation. – Ian Jun 02 '18 at 06:21
  • 1
    To put it another way, $L=L(y,v)$ and the E-L equation reads $L_y(f(x),f'(x))=d/dx[L_v(f(x),f'(x))]$. – Ian Jun 02 '18 at 06:23
  • @Ian Thank you for your quick reply.. Mmm, do you mean that L is not a functional, but a scalar function, and f and f' are not functions but variables? And could you explain what y and v are? – KYHSGeekCode Jun 05 '18 at 14:07
  • L is not a functional in this context; it takes two scalar arguments and returns a scalar. I write $y=f(x),v=f'(x)$ to emphasize that. – Ian Jun 05 '18 at 16:51
  • @Ian Maybe I would have to take a look at the The induction process of Euler-Lagrange equation – KYHSGeekCode Jun 05 '18 at 23:27
  • 1
    Not really, you can take it as a black box. The only confusing thing is that we often conflate the Lagrangian at a given point in phase space with the Lagrangian of a trajectory; the latter is the integral of the former along the trajectory. – Ian Jun 05 '18 at 23:47
  • Possible duplicates: https://math.stackexchange.com/q/580858/11127 , https://math.stackexchange.com/q/1798396/11127 and links therein. Related Phys.SE question: https://physics.stackexchange.com/q/885/2451 – Qmechanic Jun 06 '18 at 10:43
  • @Qmechanic Right it was. – KYHSGeekCode Jun 06 '18 at 11:00
  • @Ian Thanks! Something is trying to come into my mind. Also I got words that call what I wondered from the comment below. – KYHSGeekCode Jun 06 '18 at 11:03

0 Answers0