Scribd Logo
Upload

Welcome to Scribd!

  • 38 views

    LT I Differential Equations: y (T) L (Y (S) ) C(S) R(S) G(S) Q(S) R(S) C(S) Initial Condition Polynomial

    Uploaded by

    bob3173
    The document discusses Laplace transform solutions to linear time-invariant differential equations. It provides definitions and explanations of key concepts like the zero-input response, zero-state response, impulse response, and convolution integral. The zero-state response of a system is obtained by taking the inverse Laplace transform of the transfer function multiplied by the input. The impulse response is the inverse Laplace transform of the transfer function alone. Convolution allows representing the zero-state response as the convolution of the impulse response and input.

    Copyright:

    Attribution Non-Commercial (BY-NC)
    Download as PDF, TXT or read online from Scribd

    LT I Differential Equations: y (T) L (Y (S) ) C(S) R(S) G(S) Q(S) R(S) C(S) Initial Condition Polynomial

    Uploaded by

    bob3173
    38 views8 pages
    The document discusses Laplace transform solutions to linear time-invariant differential equations. It provides definitions and explanations of key concepts like the zero-input response, zero-state response, impulse response, and convolution integral. The zero-state response of a system is obtained by taking the inverse Laplace transform of the transfer function multiplied by the input. The impulse response is the inverse Laplace transform of the transfer function alone. Convolution allows representing the zero-state response as the convolution of the impulse response and input.

    Original Title

    ENAE 641

    Copyright

    © Attribution Non-Commercial (BY-NC)

    Available Formats

    PDF, TXT or read online from Scribd

    Did you find this document useful?

    Is this content inappropriate?

    The document discusses Laplace transform solutions to linear time-invariant differential equations. It provides definitions and explanations of key concepts like the zero-input response, zero-state response, impulse response, and convolution integral. The zero-state response of a system is obtained by taking the inverse Laplace transform of the transfer function multiplied by the input. The impulse response is the inverse Laplace transform of the transfer function alone. Convolution allows representing the zero-state response as the convolution of the impulse response and input.

    Copyright:

    Attribution Non-Commercial (BY-NC)
    Download as PDF, TXT or read online from Scribd
    Download as pdf or txt
    38 views8 pages

    LT I Differential Equations: y (T) L (Y (S) ) C(S) R(S) G(S) Q(S) R(S) C(S) Initial Condition Polynomial

    Uploaded by

    bob3173
    The document discusses Laplace transform solutions to linear time-invariant differential equations. It provides definitions and explanations of key concepts like the zero-input response, zero-state response, impulse response, and convolution integral. The zero-state response of a system is obtained by taking the inverse Laplace transform of the transfer function multiplied by the input. The impulse response is the inverse Laplace transform of the transfer function alone. Convolution allows representing the zero-state response as the convolution of the impulse response and input.

    Copyright:

    Attribution Non-Commercial (BY-NC)
    Download as PDF, TXT or read online from Scribd
    Download as pdf or txt
    of 8

    09-11-2008

    Laplace Transform Solutions of LT I dierential equations:


    y(t) = L
    1
    {Y (s)}
    Y (s) = G(s)U(s) +
    c(s)
    r(s)
    G(s) =
    q(s)
    r(s)
    (transfer function)
    c(s) = initial condition polynomial
    if U(s) is a rational function of s, then
    U(s) =
    a(s)
    b(s)
    where a(s), b(s) are polynomials in s.
    Then by partial fraction expansion
    Y (s) =
    n

    k=1
    A
    k
    s p
    k
    +
    L

    l=1
    B
    l
    s
    l
    where, r(p
    k
    ) = 0, for k = 1, . . . , L (p
    k
    are zeros of polynomial r(s)) and
    b(
    l
    ) = 0 for l = 1, . . . , L (
    l
    are zeros of polynomial b(s)).
    Then
    y(t) = L
    1
    {Y (s)}
    =
    n

    k=1
    A
    k
    e
    p
    k
    t
    . .
    y
    h
    (t)
    +
    L

    l=1
    B
    l
    e

    l
    t
    . .
    y
    p
    (t)
    Y (s) = G(s)U(s)
    . .

    +
    c(s)
    r(s)
    ..

    1
    where * is the response due soley to control inputs (all initial conditions are
    zero) and ** is the response of the system due soley to the I.Cs (with respect
    to no input).
    The zero input response is:
    Y
    z
    i
    (s) [y
    z
    i
    = L
    1
    {Y
    z
    i
    (s)}]
    The zero state response is:
    Y
    z
    s
    (s) = G(s)U(s)
    Lets look at the zero state response more carefully. Response of the system
    due soley to input, assuming initial conditions are zero.
    Recall the pulse and delta functions are:

    (t) =
    _
    1

    0 t
    0 otherwise
    and
    (t) = lim
    0

    (t)
    For any u(t), lets examine the piecewise constant approximation
    u(t) = u

    (t)

    k=0
    u(k)

    (t k)
    height of (t-k)=1
    1

    (t)

    (t-)

    (t-2)

    (t-3)

    2 3

    2
    u(t) = lim
    0
    u

    (t)
    =
    _

    0
    u()
    _
    lim
    0

    (t )d
    _
    =
    _

    0
    u()(t )d = u(t)
    SYSTEM (t) g (t)

    Let g

    (t) be the response of the system to the pulse function

    (t) (Assuming
    zero initial conditions). More generally, let
    g

    (t, k)
    be the response of the system to

    (t k) (pulse function starting at k).


    3
    Note formally:
    g

    (t) = g

    (t, 0)

    t
    t
    System
    k
    1

    t
    t
    System
    g (t-k)

    g (t,k)

    For a time invariant system [constant coecient dierential equation] it is


    true that:
    g

    (t, k) = g

    (t k)
    i.e. it is the original pulse response shifted in time by k units. Thus far a
    linear time invariant system:
    y

    (t) = response due to piecewise constant u

    (t)
    let:
    k
    = k then,
    d
    k
    =
    k+1

    k
    = for all k > 0
    then:
    y

    (t) =

    k=0
    u(
    k
    )g(t
    k
    )d
    k
    4
    y
    z
    s
    = lim
    0
    y

    (t)
    or:
    y
    z
    s
    (t) =
    _

    0
    g(t )u()d convolution integral
    where
    g(t) = lim
    0
    g

    (t)
    SYSTEM (t) g(t)
    g(t) is the response of the system to ideal delta functions (t), also known as
    the impulse response.
    more generally, if
    g

    (t, k) = g(t k)
    then:
    y
    z
    s
    (t) =
    _

    0
    g(t, )u()d
    g(t, ) is the response of a system at time t to an ideal impulse occuring at
    time .
    5
    t
    t
    System
    g(t,0)
    t
    t
    System
    g(t,)
    (t)
    (t-)

    g(t) = y
    z
    s
    (t)|
    u(t)=(t)
    L
    1
    {Y
    z
    s
    (s)|
    u(t)=(t)
    }
    L
    1
    {G(s)U(s)}|
    u(t)=(t)
    if U(t) = (t) then U(s)=1 and
    g(t) = L
    1
    {G(s)}
    inverse transform of the transfer function
    = L
    1
    _
    q(s)
    r(s)
    _
    = L
    1
    _
    n

    k=1

    k
    s p
    k
    _
    p
    k
    satisfy r(p
    k
    ) = 0

    k
    = [(s p
    k
    )G(s)]
    s=p
    k
    *assuming no repeated factors.
    6
    Thus:
    g(t) =
    n

    k=1

    k
    e
    p
    k
    t
    a linear combination of modes.
    y
    z
    s
    (t) =
    _

    0
    g(t )u()d
    y
    z
    s
    (t) = g(t) u(t)
    Y
    z
    s
    (s) = G(s)U(s)
    where
    G(s) = L{g(t)}
    U(s) = L{u(t)}
    L{g u}G(s)U(s)
    *Laplace turns convolution into multiplication!
    SYSTEM u(t) y(t)
    Inputs Outputs
    u(t) R
    n

    t
    u(t) =
    _

    _
    u
    1
    (t)
    u
    2
    (t)
    .
    .
    .
    u
    n
    (t)
    _

    _
    where u
    j
    (t) R
    t
    where j = 1, . . . , m
    y(t) R
    p

    t
    y(t) =
    _

    _
    y
    1
    (t)
    y
    2
    (t)
    .
    .
    .
    y
    p
    (t)
    _

    _
    y
    i
    (t) R
    t
    where i = 1, . . . , p.
    7
    1. suppose that the input u(t), t T
    0
    is known.
    2. and that initial conditions on system are known at time t
    0
    call this collection of initial conditions:
    x
    0
    x
    0
    R
    n
    where there are n initial conditions and nd solutions

    y(t), t t
    0
    .
    What formally is the state of the system?
    The states of a dynamical system are a collection of numbers, x
    k
    (t) for
    k = 1, . . . , n such that knowledge of these numbers at any time t
    0
    , together
    whith knowledge of the input u(t) for t t
    0
    allows us to compute (predict)
    the output y(t) for t t
    0
    . The most general model:
    State Equation

    x(t) =

    f(x(t), u(t), t) x(t
    0
    ) = x
    0
    Output Equation
    y(t) =

    h(x(t), u(t), t)
    8

    You might also like