0% found this document useful (0 votes)
10 views

NonLinear Lecture 1.2

This document discusses non-linear systems and phase planes. It outlines that the phase plane is a plot of two variables (x1 and x2) over time, with each point defining a trajectory. Trajectories represent solutions and are studied to understand system behavior. Examples demonstrate plotting mass-spring-damper system trajectories and finding trajectories for undamped oscillators which form ellipses in phase space. The document also notes that trajectories do not cross due to the uniqueness of their slope at each point.

Uploaded by

jiales225
Copyright
© © All Rights Reserved
Available Formats
Download as PDF, TXT or read online on Scribd
0% found this document useful (0 votes)
10 views

NonLinear Lecture 1.2

This document discusses non-linear systems and phase planes. It outlines that the phase plane is a plot of two variables (x1 and x2) over time, with each point defining a trajectory. Trajectories represent solutions and are studied to understand system behavior. Examples demonstrate plotting mass-spring-damper system trajectories and finding trajectories for undamped oscillators which form ellipses in phase space. The document also notes that trajectories do not cross due to the uniqueness of their slope at each point.

Uploaded by

jiales225
Copyright
© © All Rights Reserved
Available Formats
Download as PDF, TXT or read online on Scribd
You are on page 1/ 9

Non-Linear Systems and the Phase

Plane
Lecture 1.2

Paul White
Outline
• “Quotes” and preamble.
• State-space
• Phase plots
• Trajectories and phase portraits
• Equilibrium points
• Solutions of linear equations

2
Phase Plane
• Phase plane methods explore the behaviour of non-linear
systems in a graphical manner.
• The phase plane is a plot of x1 against x2.
• A solution of the equations defines x1(t) and x2(t), each point in
time specifies a point in the phase plane and as time evolves
that point moves along a curve within the phase plane.
• Curves that are drawn represent a solution – these curves are
called trajectories (a.k.a. phase paths or integral curves).
• By studying the nature of the trajectories we seek to
understand the behaviour of the system.

3
Example

Solution of mass-spring-damper
system: x1 is displacement, x2 is
velocity, initial condition (1,0).

4
Phase Portrait
• A plot of a range of trajectories for a particular equation in
some sense characterises all the solutions for that system.
– Note that the trajectories are curves with arrows indicating the direction
of travel of the solution.
• The trajectories are plotted for a range of different initial
conditions.
• The phase portrait is a valuable tool when trying to
“understand” the behaviour of a particular system.
• We seek methods to sketch phase portraits without explicitly
solving the system of equations.

5
Many similar solutions
plotted for different initial
conditions on the circle
x12+x22=1.

6
Finding Trajectories
• The trajectories for a system can in principle be computed
directly.
dx1
= f1 ( x1 , x2 )
dt dx1 f1 ( x1 , x2 )
⇒ =
dx2 dx2 f 2 ( x1 , x2 )
= f 2 ( x1 , x2 )
dt
• Solving this equation defines x1 as a function of x2, which
defines a trajectory.
• However, this equation is rarely any more tractable than the
original system of equations.
• Phase plane methods seek to plot trajectories using without
resorting to solving the above equation directly.
7
Example
• Undamped oscillator: 
x + ω2 x = 0
x1 = x2
x2 x =
= x1 x ⇒
x2 = −ω2 x1

∫ ∫
dx1 x
=− 22 ⇒ ω2 x1dx1 =− x2 dx2
dx 2 ω x1
⇒ ω2 x12 + x2 2 =C

• Hence the trajectories are ellipses in phase space.


• We know the solutions of this 2nd order ODE has the form:
x = A cos ( ωt + φ ) so that x = − Aω sin ( ωt + φ )
which is the parametric form of an ellipse (ω=1 is a circle).
How does the phase portrait change if we consider 
x − ω2 x = 0 8
Uniqueness
• Trajectories do not cross.
• Two separate trajectories do not cross, since at each point in
the plane the slope of the trajectory is unique, it is

dx1 f1 ( x1 , x2 )
=
dx2 f 2 ( x1 , x2 )

• However, note that when f= 1 ( x1 , x2 ) 2 ( x1 , x2 )


f= 0 then the
trajectory slopes are not defined and it is possible for
trajectories to converge to such a point.

You might also like