Title: Chaos in a Pendulum
1Chaos in a Pendulum
- Simulation,
- Experimental Characterization,
- and Prediction of
- Deterministic Periodic and Chaotic States
Eugene Donev
27th March 2002
2Galileos Apocryphal Chandelier
3The Plane Pendulum
Newtons Second Law
Torque
Important -- The restoring force is nonlinear
4The Damped and Driven Pendulum
- Swinging under the influence of
- gravity (restoring force, -)
- velocity-proportional damping (resisting force,
-) - sinusoidal driver ()
The equation of motion (relating the above
torques)
After some intense algebra
5The Damped and Driven Pendulum (continued)
Natural units for the pendulum
Then fiddling with derivatives and constants
turns this
into that
where
6The Real (Physical) Pendulum
- Getting physical
- pendulum length 2 cm eddy-current damping
(controlled by a micrometer) optical encoding of
angular position brushless slotless linear AC/DC
motor - driver sinusoidal torque frequency 0.3 - 3.0
Hz, /- 0.001 Hz output voltage 0 - 7 V - PC card and software
7The Physical Pendulum (continued)
Seen that already
The good old dimensionless parameters
But what about ?0, and b, and I, and T, and
Tcritical?
8Parameterization of the Pendulum
Determining the Natural Frequency, ?0 ?
collecting time series of angular displacement,
?, at minimum damping ? performing nonlinear
fitting of damping equation to part of the time
series ? calculating ?0 from the obtained ?
and ?, damping parameter
Now the driving frequency parameter can be
calculated and varied ?
9Parameterization of the Pendulum (continued)
b/I as a function of Micrometer Setting, S ?
collecting time series of angular displacement,
?, at various values of S ? performing
nonlinear fitting of damping equation to part of
the time series for each S ? obtaining
calibration curve for various micrometer settings
Now the damping parameter can be calculated and
varied ?
10Parameterization of the Pendulum (continued)
Normalized Torque as a function of Driving
Voltage (DC) ? pendulum turned on-edge to
eliminate gravitational restoring force ?
measuring terminal angular velocity, ?term , at
constant damping and varying drive voltage ?
obtaining calibration curve for normalized
torque, T/Tcritical , as a function of voltage
(DC)
Now the driving amplitude parameter can be
calculated and varied ?
11Tweaking the Parameters
Driving frequency (kept constant, ?d 2/3)
Micrometer damping (two different settings, q 2
and q 4)
Driving amplitude (a range of settings)
g 0.92 Period 1
g 1.44 Period 2
g 1.61 Period 3
g 1.75 Chaos
12g 0.92
13g 1.44
14g 1.61
15g 1.75
16CHAOS Aperiodic long-term behavior in
a deterministic system that exhibits sensitivity
to initial conditions.
17Phase Portrait
How do we read the Phase Portrait ?
18How to read the Phase Plane diagram
19g 0.92
20g 1.44
21g 1.61
22Pendulum Bifurcation Diagrams
Solving the differential equation of motion
numerically
becomes
I have used the fourth-order Runge-Kutta
method, with an adjustable time increment.
Solving numerically for x and for v, at small
increments of g.
23 24(No Transcript)
25g 1.44
26 27 28g 1.75
29g 1.61
30g 1.61
31g 1.75
32g 3.49
33(No Transcript)
34(No Transcript)
35(No Transcript)
36Prerequisites for Chaotic Behavior in a Physical
System
1. at least three independent dynamical variables
2. nonlinearity in the equations of motion
3. appropriate values for the parameters
37The Chaotic Pendulum (again)
1. at least three independent dynamical variables
2. nonlinearity in the equations of motion
3. appropriate values for the parameters
38(No Transcript)
39Angular Displacement time series at minimum
damping
Angle (rad)
Time (s)
?
40Data and nonlinear fit for minimum damping
Angle (rad)
Time (s)
?
41Micrometer setting 8.00 By t 4 s, the
oscillations have almost completely died
away, i.e., much damping.
Micrometer setting 11.50 By t 7 s, the
pendulum is still swinging, i.e., little damping.
?
42Normalized Damping as a function of Micrometer
Setting
?
43Normalized Torque as a function of Applied Voltage
?
44?
45(No Transcript)
46(No Transcript)
47(No Transcript)
48(No Transcript)
49(No Transcript)
50(No Transcript)