Lagrangian Methods in 1D Flashcards

1
Q

What can we use calculus of variations for?

A

Instead of guessing the path, we can find function which extremises an integral.

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
2
Q

What is the equation for the action when trying to find the minimal path?

A

A = integral from t0 to t1 of L(x(t), v(t), t) dt

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
3
Q

How can we show the path varying in the variables?

A

x(t) -> x(t) + a(t) and hence v(t) -> v(t) + a’(t)

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
4
Q

What do we do to the end points of the path and what does this mean for the Lagrangian?

A

We fix the end points: a(t0) = a(t1) = 0, so L(x, v, t) -> L(x+a, v+a’, t)

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
5
Q

Using Taylor expansion, what does the first order of the Lagrangian for very small a(t) become?

A

L(x,v,t)+a(t)dL/dx + a’(t)dL/dv

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
6
Q

What does the equation for the action become?

A

The new Lagrangian term subbed in so: integral from t0 to t1 of L(x,v,t)+a(t)dL/dx + a’(t)dL/dv dt

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
7
Q

What is 𝛿A, the added on part?

A

New part without old part of the Action equation: integral from t0 to t1 of a(t)dL/dx + a’(t)dL/dv dt

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
8
Q

What can we do to simplify 𝛿A?

A

Integrate the second term (involving a’(t)) by parts, and then use a(t0) = a(t1) = 0, so this integrated part is zero.

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
9
Q

What is 𝛿A after the integration by parts? What do we do with this?

A

𝛿A = integral from t0 to t1 of a(t)*(dL/dx - d/dt(dL/dx’))

With this, for A to be stationary, we need 𝛿A = 0, so part inside the brackets = 0.

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
10
Q

What is the Euler-Lagrange equation?

A

dL/dx - d/dt * dL/dx’ = 0

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
11
Q

How can we use the E-L equation in the mass under gravity problem?

A

Find the different terms in the E-L equation using differentiation: find that it is equal to -mg-ma = 0, so a = -g, as expected from Newtons second law.

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
12
Q

What do we start with for the alternative derivation of the E-L equation?

A

An alternative path which differs from the optimum path: x(α,t) = x(0,t)+α*Ν(t), where Ν(t) is an arbitrary function

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
13
Q

What do we do after starting the alternative derivation of the E-L equation?

A
  • Find the new Lagrangian is: L(x+αΝ, v+αΝ’, t)

- Know that for all Ν(t), path minimised for α = 0 as x(0,t) is optimum path (dA/dα at α = 0 = 0)

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
14
Q

What is the integral from t0 to t1 of L(t,x,v) dt equal to?

A

= f(x,v)

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
15
Q

How do we find dA/dα for the alternative derivation of the E-L equation?

A
  • dA/dα = d/dα*integral from t0 to t1 of L dt = df/dα

- Use chain rule so df/dα = df/dx * dx/dα + df/dv * dv/dα

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
16
Q

What do we do with df/dα after using the chain rule?

A
  • Sub in the Lagrangian integral again and rearrange so integral is on outside of brackets
  • Use equations for x and v to differentiate
  • Integrate by parts
17
Q

What do we get after integrating by parts?

A
  • dA/dα = integral from t0 to t1 of Ν*(dL/dx - d/dt * dL/dv)
  • if dA/dα at α = 0 = 0 for any Ν(t), we must have: dL/dx - d/dt * dL/dv = 0
  • The E-L equation
18
Q

How do we use the E-L equation in a 1D potential, potential energy V(x)? (find Newtons second law)

A
  • L = T-V = 1/2 * m*v^2 - V(x)
  • Apply E-L: dL/dx = -dV/dx = force due to potential, d/dt(dL/dv) = dρ/dt
  • F = dρ/dt = Newtons second Law
19
Q

For a swinging pendulum, what do we make the Lagrangian a function of and why?

A

L(θ, θ’), since the length is fixed, we can describe the position using only θ, don’t need x,y.

20
Q

What is the potential energy V equal to for the pendulum?

A

V = const - mgLcosθ

21
Q

What is the kinetic energy T equal to for the pendulum?

A

1/2 * m*v^2 = 1/2 * m(Lθ’)^2, where v = (Lθ’)^2

22
Q

What do we do once we have found the Lagrangian for the simple pendulum?

A

Sub it into the E-L equation and find the differentials, then rearrange to find an equation for simple harmonic motion: θ’’ = -g/L * θ

23
Q

For a rollercoaster, what can we use instead of x,y and why?

A

Can use arc length s(t), since it is a fixed path and x&y aren’t independent.

24
Q

What is the Lagrangian equal to for the roller coaster?

A

L = 1/2 * m*s’^2 - mgh(s) = L(s,s’)

25
Q

What do we do with the Lagrangian for the roller coaster problem?

A

Sub it into the E-L equation, and find s’’ = -g * dh/dm which is the equation of motion, so if we know h(s), we get s(t).

26
Q

What is the generalised version of the E-L equation?

A

dL/dq 0 d/dt * dL/dq’ = 0, where q(t) is a generalised coordinate and may be x(t), θ(t) etc, and q’ is the related velocity.