Strange invocation of Taylor series

Join the discussion
Ask a follow-up here, or get your own question answered by working scientists, mathematicians and engineers — people, not an autocomplete.
Real named experts · corrections over time · the nuance an AI answer skips
6 replies · 1K views
noahcharris
Messages
21
Reaction score
0
Hi all,

I was working through a chapter on Lagrangians when I cam across this:

"Using a Taylor expansion, the potential can be approximated as
## V(x+ \epsilon) \approx V(x)+\epsilon \frac{dV}{dx} ##"

Now this looks nothing like any taylor expansion I've seen before. I'm used to

## f(x) = \sum\frac{f^{(i)}(a)(x-a)^i}{i!} ##

What am I missing here?
 
Physics news on Phys.org
noahcharris said:
Hi all,

I was working through a chapter on Lagrangians when I cam across this:

"Using a Taylor expansion, the potential can be approximated as
## V(x+ \epsilon) \approx V(x)+\epsilon \frac{dV}{dx} ##"

Now this looks nothing like any taylor expansion I've seen before. I'm used to

## f(x) = \sum\frac{f^{(i)}(a)(x-a)^i}{i!} ##

What am I missing here?
You just need to expand the definition of the Taylor series over the first couple of terms. Remember to account for x + ε as the argument of V(x).
 
  • Like
Likes   Reactions: noahcharris
What are the first two terms with a = x + ξ ?
 
  • Like
Likes   Reactions: noahcharris
paisiello2 said:
What are the first two terms with a = x + ξ ?

Looks like it's ## f(x) \approx f(x + \epsilon) - f^1(x + \epsilon)\epsilon ##

Oh, ok, I see now. So the ##\frac{dV}{dx}## is really a ##\frac{dV(x + \epsilon)}{dx}## ??
 
You don't even need to use a Taylor series. Just use the definition of a derivative:
[tex]\frac{dV}{dx}= \lim_{\epsilon \to 0} \frac{V(x+\epsilon)-V(x)}{\epsilon}[/tex]
Therefore, for smaller and smaller [itex]\epsilon[/itex], the above becomes a better and better approximation. The expression from your first post just comes from taking the above equation, dropping the limit, and isolating [itex]V(x+\epsilon)[/itex].
 
  • Like
Likes   Reactions: noahcharris
noahcharris said:
Looks like it's ## f(x) \approx f(x + \epsilon) - f^1(x + \epsilon)\epsilon ##

Oh, ok, I see now. So the ##\frac{dV}{dx}## is really a ##\frac{dV(x + \epsilon)}{dx}## ??
Sorry, do what SteamKing suggested but also take a=x as where the approximation is taken from. That should give the same answer.
 
Ok I finally figured it out. I was wrong in my previous comment. ## x \mapsto x + \epsilon ## and ## a = x ##
Thanks everyone.