0
$\begingroup$

I have read in my book that if $\frac{\partial L}{\partial t}=0$, then the quantity $ L-\frac{\partial L}{\partial \dot{q}} \dot{q} $ is conserved, and we call it the energy of the system.

But if this is the energy of the system there is something I misunderstand with the fact that adding a term like $\frac{d}{dt} f(q,t)$ doesn't change the equations of motion.

Indeed, we could write :

$$L'=L+\alpha t$$

And $L'$ and $L$ describe the same physics of the system because $t$ can be written as $\frac{d (\frac{t^2}{2})}{dt} $.

This second lagrangian $L'$ explicitly depends on time and then the energy is not conserved, if we do the calculation we would have this equality :

$$ L'-\frac{\partial L'}{\partial \dot{q}} \dot{q}=\alpha t+b$$

So it looks like the energy is not conserved in time for the new lagrangian.

So there is something I misunderstand...

$\endgroup$
  • $\begingroup$ $b$ is conserved and that is defined as the energy. The LHS is not the energy, since $L'$ does not satisfy to the condition. This is a special case of Noether's theorem. $\endgroup$ – Phoenix87 Dec 27 '16 at 14:23
  • $\begingroup$ What do you call "LHS" ? Im a beginner in lagrangian mechanics. If $b$ is not the energy, how can we define it in a general case ? In the book I read they define the energy as $-L+\frac{\partial L}{\partial \dot{q}}\dot{q}$ so they are wrong in a general case ? $\endgroup$ – StarBucK Dec 27 '16 at 14:45
1
$\begingroup$
  1. No explicit time dependence of the Lagrangian is just a simplifying sufficient condition for Noether's theorem. It is not necessary. Recall that Noether's Theorem (in its original form) concerns a quasi-symmetry of the action $S$, cf. this Phys.SE answer.

  2. If the Lagrangian is of the form $$\tag{1} L(q,\dot{q},t)~=~ L_0(q,\dot{q}) +\frac{dF(q,t)}{dt},$$ where $L_0(q,\dot{q})$ has no explicit time dependence, then a notion of energy is $$\tag{2} \dot{q}^i\frac{\partial L_0}{\partial \dot{q}^i }-L_0.$$ It is possible to view the energy (2) as a Noether charge associated with time translations by modifying the techniques of this Phys.SE post.

$\endgroup$

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.