Location via proxy:   [ UP ]  
[Report a bug]   [Manage cookies]                
Scribd Logo
Upload

Welcome to Scribd!

  • 89 views

    EE132 Lab1 OL Vs CL

    Uploaded by

    thinkberry22
    This document describes an experiment comparing open-loop and closed-loop control of a DC motor. The motor is modeled using state space equations, and disturbances, modeling errors, and control strategies are analyzed both theoretically before the experiment and through simulation. In the experiment, an open-loop controller is first designed and tested in simulation under nominal and off-nominal conditions. Tracking error increases with disturbances. Then, a closed-loop PI controller is designed and shown in simulation to reduce tracking error even in the presence of disturbances and modeling errors.

    Copyright:

    © All Rights Reserved
    Download as PDF, TXT or read online from Scribd

    EE132 Lab1 OL Vs CL

    Uploaded by

    thinkberry22
    89 views3 pages
    This document describes an experiment comparing open-loop and closed-loop control of a DC motor. The motor is modeled using state space equations, and disturbances, modeling errors, and control strategies are analyzed both theoretically before the experiment and through simulation. In the experiment, an open-loop controller is first designed and tested in simulation under nominal and off-nominal conditions. Tracking error increases with disturbances. Then, a closed-loop PI controller is designed and shown in simulation to reduce tracking error even in the presence of disturbances and modeling errors.

    Original Description:

    Hhgg

    Original Title

    EE132 Lab1 OL vs CL

    Copyright

    © © All Rights Reserved

    Available Formats

    PDF, TXT or read online from Scribd

    Did you find this document useful?

    Is this content inappropriate?

    This document describes an experiment comparing open-loop and closed-loop control of a DC motor. The motor is modeled using state space equations, and disturbances, modeling errors, and control strategies are analyzed both theoretically before the experiment and through simulation. In the experiment, an open-loop controller is first designed and tested in simulation under nominal and off-nominal conditions. Tracking error increases with disturbances. Then, a closed-loop PI controller is designed and shown in simulation to reduce tracking error even in the presence of disturbances and modeling errors.

    Copyright:

    © All Rights Reserved
    Download as PDF, TXT or read online from Scribd
    Download as pdf or txt
    89 views3 pages

    EE132 Lab1 OL Vs CL

    Uploaded by

    thinkberry22
    This document describes an experiment comparing open-loop and closed-loop control of a DC motor. The motor is modeled using state space equations, and disturbances, modeling errors, and control strategies are analyzed both theoretically before the experiment and through simulation. In the experiment, an open-loop controller is first designed and tested in simulation under nominal and off-nominal conditions. Tracking error increases with disturbances. Then, a closed-loop PI controller is designed and shown in simulation to reduce tracking error even in the presence of disturbances and modeling errors.

    Copyright:

    © All Rights Reserved
    Download as PDF, TXT or read online from Scribd
    Download as pdf or txt
    of 3

    1

    EE 132 Lab 1: Open vs Closed Loop


    I. OBJECTIVE
    To compare the performance of open and closed loop controllers, applied to a simple plant, in the presence of
    disturbances or modeling error.
    II. EQUIPMENT
    Matlab and Simulink
    III. BACKGROUND
    Consider a DC motor shown in Fig. 1. The schematic diagram for this system is shown in Fig. 2, where R is the
    electric resistance, L is the electric inductance, is the angle of the shaft, =

    is the angular rate of the shaft, T
    is the motor torque, V is the input voltage, J is the moment of inertia of the rotor, b is the coefcient of friction
    of the mechanical system. For a DC motor, we have the following relations
    T = Ki (1)
    e = K (2)
    where i is the armature current, e is the back electromotive force, and K is the electromotive force constant. Let
    d(t) represent an unknown load torque applied to the shaft. For example, if this is a grinding application, then the
    torque d would be determined by the force applied to the grinding disk by the user.
    The state space model of this system is
    J + b = Ki + d (3)
    L
    di
    dt
    + Ri = V K (4)
    From (4), we have
    i =
    L
    R
    di
    dt
    +
    V
    R

    K
    R
    .
    We will assume that the motor has been selected such that R >> L, then L/R 0 and
    i =
    V
    R

    K
    R
    .
    Substitute i into (3), we have
    J +

    b +
    K
    2
    R

    =
    K
    R
    V + d. (5)
    Selecting the state variable as x = , equation (5) can be written as
    x =

    bR + K
    2
    JR

    x +
    K
    JR
    V +
    1
    J
    d. (6)
    If the nominal parameter values are as follows
    b = 0.01N m s, J = 1.0 kg m
    2
    , K = 0.1
    N m
    A
    , R = 1.0
    then, equation (6) becomes
    x = 0.02x + 0.1V + d. (7)
    In (7), V is the control input, d can be considered as a disturbance, x is the angular velocity of the shaft. All units
    in this lab are international units.
    2
    Fig. 1. DC motor Fig. 2. DC motor System
    In practice, the nominal parameter values are not known exactly and may change with time. So, a more realistic
    model than (7) is
    x = (0.02 + a)x + (0.1 + b)V + d (8)
    where a and b are uncertain constants terms.
    For the prelab portion of this assignment, we will analyze the effects of a, b, and d.
    For the in lab simulation portion of this assignment, we will only be concerned with the disturbance d.
    System (7) is called the nominal system corresponding to (8). The nominal system can be obtained by letting
    a = b = 0 in (8).
    IV. PRELAB (COMPLETE THIS SECTION PRIOR TO LAB.)
    In this section, we assume that the reference state x
    c
    (t) = 1u(t), where u(t) is a step function, i.e., u(t) = 1
    for all t 0 and u(t) = 0 for all t < 0.
    1) For system (7), design an open loop controller
    V (t) = F
    1
    x
    c
    (t) (9)
    where F
    1
    is a constant.
    a) Find F
    1
    such that lim
    t
    (x(t) x
    c
    (t)) = 0 when d = 0, a = 0, and b = 0.
    b) Where is the pole of the controlled system when d = a = b = 0? Is the pole location affected by the
    choice of the control law parameter F
    1
    ?
    c) What is the settling time of the closed-loop system when d = a = b = 0 (i.e., three times the time
    constant when d = a = b = 0)?
    d) Find an expression for the tracking error x
    e
    = (x x
    c
    ) as a function of d, a, and b.
    2) For system (8), dene the closed loop controller
    V (t) = P(x
    c
    (t) x(t)) + I

    t
    0
    (x
    c
    () x())d (10)
    where P and I are constant real numbers.
    a) Substitute eqn. (10) int eqn. (7) with d = 0, transform to the frequency domain, and simplify, to nd
    the closed loop transfer function from X
    c
    (s) to X(s).
    b) Find P and I such that the closed loop transfer function from X
    c
    (s) to X(s) has poles at 0.1 j0.05
    when d = a = b = 0.
    3
    i) To nd the desired characteristic equation (i.e., transfer function denominator), nd a polynomial
    that has the specied roots.
    ii) Find P and I so that the denominator of the closed loop transfer function matches the desired
    characteristic equation.
    c) What is the settling time of the closed-loop system when d = a = b = 0 (i.e., three times the 1/||,
    where is the real portion of the pole when d = a = b = 0.)
    d) What is the tracking error x
    e
    = (xx
    c
    ) for constant values of d, a, and b, assuming the system remains
    stable.
    V. SIMULATION PROCEDURE
    Please download the le Lab1.mdl from iLearn. Open the Simulink library in MATLAB. The simulation le
    can be opened in the Simulink library. Finish the simulation implementation by the following steps.
    1) The simulation implementation includes portions of the control law implementation. Check the implementation
    of the plant model. Make sure that you agree, then proceed to the following steps.
    2) Implement the open loop controller. Let x
    c
    be a step input at time zero with a magnitude of 1.0. Set the
    length of the time interval to be 600 seconds for this case.
    a) Simulate the response of system (8) with control law (9) when d = a = b = 0. In this step your goal is
    to ensure that your prelab design and simulation implementation are correct (i.e., The steady state error
    should be zero. Steady state should be approximately achieved in three time constants.)
    b) Simulate the response of the open loop system for d = 0.1 and a = b = 0.
    c) Show the results of the three simulations on one graph. Every time when you launch the Simulink le,
    the time array and corresponding state array will be saved as Data in the workspace of MATLAB.
    The time array can be obtained by using Data.time. The state array can be obtained by using
    Data.signals.values. Then you can use your knowledge of MATLAB to plot what gure that is asked.
    Ask the TA if you do not understand how to do this. Clearly label each graph. In your lab report, discuss
    the effect of the off-nominal effects on the system response. Compare the simulated responses with the
    prelab analysis.
    d) Remove the open loop controller.
    3) Implement the closed loop controller. Again, let x
    c
    be a step input at time zero with a magnitude of 1.0. Set
    the length of the time interval to be 120 seconds for this case.
    a) Simulate the response of system (8) with control law (10) when d = a = b = 0. In this step your goal
    is to ensure that your prelab design and simulation implementation are correct (i.e., The steady state
    error should be zero. Steady state should be approximately achieved in three time constants.)
    b) Simulate the response of system (8) with control law (10) for d = 0.1 and a = b = 0.
    c) Show the results of the three simulations on one graph. Ask the TA if you do not understand how to
    do this. Clearly label each graph. In your lab report, discuss the effect of the off-nominal effects on the
    system response. Compare the simulated responses with the prelab analysis.
    Congratulations. You have implemented your rst closed loop control law. The simulation allows an easy means to
    analyze the affect of changing the other model error terms. Investigate them if you are interested.
    FIGURES - Throughout this class, all gures must be properly labeled
    Each axis must have a variable name with units (e.g. tracking error, m/s).
    The gure must have name (e.g., Figure 1).
    Each gure must be mentioned by name and discussed in the text (e.g. Figure 1 shows that ...). The discussion
    must state any conclusion that the students expects the reader to draw from the gure.
    If a graph contains more than one curve, each curve should have a different line type. Each line type must
    be dened. Each curve must be discussed. For example, Figure 1 shows that for d = 0.0 (solid curve) the
    tracking error achieves a steady state value of ... in ... seconds and for d = 0.1 (dashed curve) the tracking
    error achieves a steady state value of ... in ... seconds.
    This direction will not be repeated in future labs, but is expected to be followed.

    You might also like