UCK303E Automatic Control
Lecture 3 and 4 : Time Domain
 Analysis of Dynamic Systems
  and Introduction to Laplace
        Transformation
                       Previous Lecture
• Dynamic modeling of mechanical and electrical
  (analog) devices
   – based on basic physical principles (Newton’s law, Kirchhoff’s
     current and voltage law…).
   – using Mass/Inductance (Inertia), Spring/Capacitance
     (Energy-storage), Damper/Resistance (Energy-dissipation)
     elements.
   – resulting in second order ODEs, higher order ODEs, system
     of differential equations.
• Two interesting sub-topics:
   – state-space representation of dynamic systems
   – linearization of non-linear equations of motions
                                III. Suspension of a Car
Output of interest: Vertical position of chassis
Plant: Vehicle chassis (including tires), some actuators, and their dynamics
Inputs: Altitude of car chassis with bumps in the road
                        Suspension of a Car
                         Applying Newton’s Law to each mass
                         When we rearrange these formulas:
Free-body diagram for
  suspension system                                           4
        IV. Rotational Pendulum
Follow the same steps :
Small theta assumption :
                  : natural frequency…
                                         5
                             In this lecture
• We explore the time-domain understanding of the
  transformation by solving a second-order ODE
      • Explicit time-domain solution by finding the homogenous and
        forced response.
      • Modes of the system and the homogenous and forced response
          – Homogenous Solution => Basic Stability Understanding (complex plane
            plots of the natural frequencies).
          – Forced Solution => Frequency Response (gain and phase plots).
• We explore the both the stability notion and the forced
  response characteristics via examples.
                    We will continue with…
• Laplace Transformation (Frequency Domain).
• Dirac Delta, Convolution
   => Inherit tie between time domain and frequency domain…
• Transfer Function representation of dynamic systems (will
  continue in the next lecture)
• And we enter into automatic control systems via analyzing the
  stability and the time-response of second order dynamic systems
  (will continue in the next lecture)…
                              Lecture Outline
• Later we will look at linear time-invariant differential
  equations.
   – Higher-order ODEs and system of differential equations
       • State-space representation : set of first order linear differential
         equations
      Textbook References and Reading Assignment
• In this lecture we will cover the topic mainly from Ogata and
  FPE , and we will use some excerpts from Kuo, Nise, Goodwin,
  Dorf, Oishi (UCB) – thank you!!!.
• Did you do your last week’s reading assignment?
   – FPE : Chapter 2 or
   – Kuo : Chapter 4– relevant portions
• Textbook follow-up (this week’s reading assignment) :
   – Ogata : Chapter 2 main text + Example problem and solutions
   – FPE : Chapter 2
   – Kuo : Chapter 2
Solving an ODE
                     Trailer Example
The car comes to a stop : how is the behavior of the trailer?
                          Almost like a wall when it stops…
Let’s denote b= Bh+Bt and k=Kh and analyze the system…
                        Mass-spring-damper system
Step 1 : Mass-spring-damper (friction) system   Step 2 : Free-body diagram
   Step 3 : Equations of Motion                                Denote u(t):=f(t) as the
                                                               input to the system
Let’s look at the solution of this ODE
                       Complex Number
 Real and Imaginary
 Portion
 Carry “Information”
real part   Complex part
              Recall Complex Numbers and Euler’s Theorem
Real and Imaginary Portion
Carry “Information”
Homogenous solution on the complex plane
Focusing on the complex root case
Homogenous solution on the complex plane
                                   Initial State Response
                         displacement
            Complex pair                                    time
             Real Pair
Response Types in Complex Plane
           Stable and unstable regions in the complex plane
S domain
Forced Response
           Uo=Fo (recall slide 12)
The solution
Generic Second Order System
                     Generic Input – Output Set
                    Assuming that the initial conditions are all zero
                                                             Roots of this are called “the poles”
                                                             of the system….
G(s) is called the “transfer function”
Generic Second Order System
            Homogenous Response
Frequency Domain     Time Domain
Ties to Future Lectures…
Forced Response in the frequency plane
                   Let’s just wait for another second on
                   What the wn and ksi means…
                   This is a very important characteristic
                   of linear systems.
Signals and Systems Point of View
Into the depths of Laplace
      Transformation
Laplace Transformation Fundamentals
Laplace, himself…
Laplace Transformation
Laplace and Inverse Laplace Transformation
        Inverse Laplace Transformation:
Note : Laplace Transformation Existence
Example I : Exponential
Example II : Step
Example III : Sinusoidal
Laplace Transform Properties
Laplace transform theorems
Important Laplace Transformation Pairs
Second Order Systems
Laplace transform table
     From time domain
   to frequency domain -
The convolution perspective...
State-Space Formulation
       Two key properties of linear constant systems
• The response of a linear
  constant (i. e. not time
  varying system
  parameters such as k, b,
  m…) system
   – Obeys the principle of
                                For the sake of this discussion
     superposition              Imagine this….
   – Can be expressed as the
     convolution of the input
     with the unit impulse
     response of the system
                 Superposition Principle
• General Signals –
  decompose it into a sum
  of elementary signals :
   – Impulse, exponential, …
In Class Example :
                            Dirac Delta Function
 Physical Understanding: Impulse
  Norm bounded ~ “Finite Energy”
If a function f(x) is continuous at x :
  Notice that the Dirac Delta is acting like a “sampling” operator at x=0 !
             First step to convolution : Sampling
                                                                                   f(t)
• For an arbitrary T and
  again assuming f(t) is
  continuous at t=T :
Notice that the Dirac Delta is acting like a “time shift” sampling operator now!
  How about making T not a constant, but the free time variable t?
                                   Convolution
Representing f(t) as a sum of impulses :
   Example : How about u(t)? (i.e. an input signal…)
       H(t,tau) : Impulse Response of a Dynamic System
                                       : identical
                                                     convolution
In Class Example :                                     Use Dirac Delta
                                                       As u
    Let’s find the impulse response…
       Complex Numbers and Euler’s Theorem
Real and Imaginary Portion
Carry “Information”
                       Transfer Function
Notice how the analysis turned into a simple algebraic multiplication…
Time domain vs. Frequency domain
Laplace Transformation Fundamentals
Laplace Transformation
Forced Response Characteristics
Trailer Example
  Almost like a wall when it stops…
               Mass-spring-damper system
      IF y2 –y1 = y
Mass-spring-damper (friction) system   Free-body diagram
                      State diagram
Example to come…
 Second-
   order
 systems’
   step
responses
  Second-
   order
    step
 response
components
 generated
     by
  complex
   poles
    Step
responses
     for
 second-
   order
  system
 damping
   cases
 Second-
   order
response
   as a
 function
    of
damping
   ratio
Second-order
underdamped
responses for
damping ratio
   values
Second-order
underdamped
  response
specifications
 Percent
overshoot
    vs.
 damping
   ratio
Normalized rise
    time vs.
    damping
   ratio for a
 second-order
 underdamped
   response
Into the depths of Laplace
      Transformation
        Example I : Mechanical Systems
Force-velocity, force-displacement, and impedance
translational relationships for springs, viscous dampers, and
mass
        Example II : Mechanical Systems
Torque-angular velocity, torque-angular displacement,
and impedance rotational relationships for springs, viscous
dampers, and inertia
            Example III : Electrical Systems
Voltage-current, voltage-charge, and impedance relationships
for capacitors, resistors, and inductors
                                                V(s)/q(s)   I(s)/V(s)
Basic time-domain test signals for control systems
Step function                          Ramp-function
                  Parabolic function