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

Class.mech & chaos theory

  1. Aug 8, 2005 #1
    Was wondering if anyone can enlighten me on what people research, when they study classical mechanics & chaos theory? Simple systems? many particle systems? do they code?
  2. jcsd
  3. Aug 8, 2005 #2
    For starters, they try to find out the solutions (well, properties and the behaviour of those solutions without actually acquiring the mathematical expression for the solution) of differential equations without actually solving the diff equations. I do not know if you have ever heard of BIFURCATIONS and Bifurcation Theory
    I had an intro course on this in my second year at the university.

  4. Aug 8, 2005 #3

    Claude Bile

    User Avatar
    Science Advisor

    Any system with feedback will behave chaotically (basically, since positive feedback acts as a sort of memory).

    Marlon mentioned a bottom up approach, starting with the equations, but many experimentalists study chaos from the top down, that is, making direct observations of chaos and attempting to quantify their observations and relate those quantities back to system parameters.

    Chaos is commonly characterised by defining the dimension of the chaos and the Lyapunov exponent (the rate of divergence of two nearly identical trajectories in the phase space of the system). These two quantities are extremely difficult to calculate and require elaborate computations to do so.

    Studying a chaotic system essentially involves the calculation of these two parameters. Chaotic systems usually have several regions of chaos, noticable changes in the behaviour of the system, depending on the amount of positive feedback. By knowing what parameters give what type of chaos, the chaos in a system can be actively controlled.

  5. Aug 8, 2005 #4
    coo, thx for the replies...is chaotic theory/dynamicalsystems/bifurcation theory(marlon, yeah i know bifurcation, funny how it applies to psych) (ie using the billiard tables system in 3D environment?) used in QM or AP alot or are there very few researchers who use it?
    Last edited: Aug 8, 2005
  6. Aug 9, 2005 #5

    Claude Bile

    User Avatar
    Science Advisor

    AP = Applied Physics?

    Well, if you do mean Applied Physics, then yes, chaos is widely studied. We have people at my university working on optical chaos and chaos in semiconductor lasers.

    The amount of papers published on these topic was enough for SPIE to release a collection of papers in their milestone series, so yes I would say it is fairly widespread.

    The thing about chaos is that it is not system specific, it turns up in just about every field of physics.

  7. Aug 9, 2005 #6
  8. Aug 10, 2005 #7

    Claude Bile

    User Avatar
    Science Advisor

    Ah, yes it does turn up in Astrophysics. Here are two examples I have come across.

    - Bistability in organic molecular clouds.
    - When one includes the gas giants when analysing the motion of the solar system, the small effect of the gas giants can induce a choatic wobble in the earths orbit that may be responsible for long term climate change. I think there was an article in New Scientist on this topic a while ago.

  9. Aug 11, 2005 #8
    Well you are talking about Picard's iteration method,and similar things!
    But they have limited application,they can't approximate every diff. eqn.
  10. Aug 11, 2005 #9
    They can study anything you listed. The simplest mechanics problem which has
    a chaos theory application is which way a pencil will fall when balanced on its point.
  11. Aug 11, 2005 #10


    User Avatar
    Science Advisor
    Homework Helper
    Gold Member
    Dearly Missed

    If I remember correctly, the mathematical study of chaos was brought into the forefront of applied maths when it was pointed out that a typical set of diff.eqs. used in meteorology was inherently chaotic.
    Meteorology is a field dominated by classical physics modelling (and no discernible improvement would be found if you were to try a QM or relativistic approach).
    Last edited: Aug 11, 2005
  12. Aug 12, 2005 #11
    Yes, i once studied the application of numerical calculus in meteorology. More specifically the contribution of Lorentz. This was just an example, in my course, of how this stuff can be used in real life. Look at page 9 and chapter 1.4 of this site

Share this great discussion with others via Reddit, Google+, Twitter, or Facebook