Dismiss Notice
Join Physics Forums Today!
The friendliest, high quality science and math community on the planet! Everyone who loves science is here!

What is the advantage of Hamilton's canonical equations?

  1. May 8, 2014 #1

    I would like to know that in what circumstances Hamilton's canonical equations are superior to the Lagrange-equations of the second kind. We know that every second order equation can be rewritten as a system of first order equations.


  2. jcsd
  3. May 8, 2014 #2


    User Avatar
    Science Advisor
    Gold Member

    There are several advantages, but I don't think they are related to the constraints:

    1. Hamilton's equations are linear and first order PDEs; Lagrange's equations non-linear and second order.
    2. The solutions to Hamilton's equations exist in phase space, and have very nice properties; Lagrange's equations exist in a different space, and the solutions may have some not-so nice properties.
    3. The linearity of Hamilton's equations comes at a price: twice as many equations. But they can be easily solved by numerical integration when there are no analytic solutions ... which is most of the time.

    See http://en.wikipedia.org/wiki/Symplectic_manifold for some discussion
  4. May 8, 2014 #3
    1. Why are they linear PDEs? See http://encyclopedia2.thefreedictionary.com/Hamilton's+Canonical+Equations+of+Motion. I do not refer to the Hamilton-Jacobian equation: http://en.wikipedia.org/wiki/Hamilton–Jacobi_equation
    2. So you mean that it has nice properties when we use apply stability analysis?
    3. It is true since ODE solvers need first order equations, but rewriting Lagrange-equations as first order equations will do the same, doesn't it?
  5. May 8, 2014 #4


    User Avatar
    Science Advisor
    Gold Member

    Hamilton's equations are partial differential equations ... your reference has suffered during translation! You can see the partial derivative. They are also linear in the variables; they are also coupled.

    The Lagrange equations are non-linear. Good luck with your hope to convert them to first order ... this is only guaranteed for linear systems. But if you perform a Legendre transform of q'->P then you get the pair of first order Hamilton equations.

    For the nice properties see any advanced text on Hamiltonian mechanics ... I'm away from my books.
  6. May 8, 2014 #5


    User Avatar
    Science Advisor
    Homework Helper

    I don't find Hamilton's equations superior in any way. If you can solve a system's Hamilton's equations, you could have solved the (Euler-)Lagrange ones as well.

    OTOH, the Hamiltonian formalism as a whole is relevant for quantization which puts in a greater emphasis in teaching it than the one which would be put on the Lagrangian formalism.
  7. May 9, 2014 #6


    User Avatar
    Science Advisor
    Gold Member

    This page summarizes the nicest property of Hamiltonian vs Lagrangian solution space:

    http://books.google.de/books?id=ebT...in phase space lagrangian hamiltonian&f=false

    Note that all of the "equation solvers" for mechanical systems (e.g., FEM, Solid Works and its brethren, etc) use the Hamiltonian form ... the numerical solutions are more stable, converge quicker, and the phase space is simpler - even though it has twice as many dimensions.
Know someone interested in this topic? Share this thread via Reddit, Google+, Twitter, or Facebook

Similar Discussions: What is the advantage of Hamilton's canonical equations?