Scribd
Upload
291 views

State Space To Transfer Function Examples PDF

This document provides two examples of deriving transfer functions from state-space representations of linear, time-invariant, continuous-time systems. In both third-order systems, the transfer function is obtained by taking the Laplace transform of the state equations, solving for the state vector X(s), and substituting into the output equation. The resulting transfer function expressions relate the input and output signals in the transform domain. The examples demonstrate that the transfer function depends on the system matrices and can reveal the system's canonical form.

Uploaded by

Gerald Aguilar
Copyright
© © All Rights Reserved
Available Formats
Download as PDF, TXT or read online on Scribd
291 views

State Space To Transfer Function Examples PDF

This document provides two examples of deriving transfer functions from state-space representations of linear, time-invariant, continuous-time systems. In both third-order systems, the transfer function is obtained by taking the Laplace transform of the state equations, solving for the state vector X(s), and substituting into the output equation. The resulting transfer function expressions relate the input and output signals in the transform domain. The examples demonstrate that the transfer function depends on the system matrices and can reveal the system's canonical form.

Uploaded by

Gerald Aguilar
Copyright
© © All Rights Reserved
Available Formats
Download as PDF, TXT or read online on Scribd
You are on page 1/ 2

1

State Space to Transfer Function Examples

A. Introduction
For a linear, time-invariant, continuous-time system, the state and output equations are

x(t) = Ax(t) + Bu(t), y(t) = Cx(t) + Du(t) (1)
where x <n is the state vector, u <r is the input (control) vector, y <m is the output vector, and {A, B, C, D} are
matrices of appropriate dimensions.
To determine the expression for the transfer function or transfer matrix, the Laplace Transforms of the above equations are
taken. If r = m = 1the single-input, single-out casethe result of this operation is a single transfer function. If there are
multiple inputs and/or multiple outputs, the result is an m r matrix of transfer functions. Taking the Laplace Transform of
the state equation yields

sX(s) x(0) = AX(s) + BU (s) (2)


where x(0) is the initial value of the state vector in the time domain. Collecting terms in X(s) and solving for X(s) gives us
the following:

(sIn A) X(s) = x(0) + BU (s) (3)


1 1
X(s) = (sIn A) x(0) + (sIn A) BU (s) (4)
The expression in (4) implies that the solution to the state equation in the transform domain has a term related to the initial
condition x(0) and a term related to the control input U (s). It will be seen later that the same holds true in the time domain.
The transfer function (or transfer matrix) is formed by assuming that the initial condition is x(0) = 0 and substituting the
remaining expression for X(s) into the output equation from (1). This yields
h i
Y (s) = C (sIn A)1 BU (s) + DU (s) = C (sIn A)1 B + D U (s) (5)
The transfer function is the relation between the transform of the input signal and the transform of the output signal. Therefore,
the transfer function (or transfer matrix) is given by

1 CAdj (sIn A) B CAdj (sIn A) B + |sIn A| D


H(s) = C (sIn A) B+D = +D = (6)
|sIn A| |sIn A|
where Adj (sIn A) is the adjoint matrix associated with (sIn A) . The term |sIn A| is an nth polynomial, the roots
of which are the poles of the transfer function. If D = 0, the numerator of H(s) in the single-input, single-output case is a
polynomial of degree n 1 or less. In this case the transfer function is strictly propermore poles than zeros. If D 6= 0, the
numerator and denominator polynomials of H(s) are both of degree n, and the transfer function is bipropersame number
of poles and zeros.

B. Example 1
The matrices of a third-order system are given by

0 1 0 0
A= 0 0 1 , B = 0 , C= 15 5 0 , D=0 (7)
8 14 7 1
Therefore, n = 3 and r = m = 1, so a single 3rd -order transfer function describes the input-output relationship for this system
in the transform domain. Going through the process step-by-step to derive the transfer function gives the following expressions.

s 0 0 0 1 0 s 1 0
sI3 A = 0 s 0 0 0 1 = 0 s 1 (8)
0 0 s 8 14 7 8 14 s + 7
The matrix of minors and matrix of cofactors are
2 2
s + 7s + 14 8 8s s + 7s + 14 8 8s
M = (s + 7) s (s + 7) 14s + 8 , Cof = (s + 7) s (s + 7) (14s + 8) (9)
1 s s2 1 s s2
2

and the adjoint matrix is


2
s + 7s + 14 (s + 7) 1
Adj (sIn A) = 8 s (s + 7) s (10)
8s (14s + 8) s2
The determinant of (sIn A) is

|sIn A| = s s2 + 7s + 14 + 1 (8) + 0 (8s) = s3 + 7s2 + 14s + 8 (11)
The steps in the calculation of the numerator are

1
Adj (sIn A) B = s , CAdj (sIn A) B = 15 + 5s (12)
s2
so the transfer function is
5s + 15 5 (s + 3)
H (s) = = (13)
s3 2
+ 7s + 14s + 8 (s + 1) (s + 2) (s + 4)
Inspection of the state and output equations in (1) show that the state space system is in controllable canonical form, so the
transfer function could have been written down directly from the entries in the state space matrices. This would not be the
case if the state space matrices were not in a canonical form.

C. Example 2
The matrices of another third-order system are given by

7 1 0 20
A = 14 0 1 , B = 125 , C= 1 0 0 , D=5 (14)
8 0 0 185
Again, n = 3 and r = m = 1, so a single 3rd -order transfer function describes the input-output relationship for this system
in the transform domain. The {A, B, C} matrices are in observable canonical form in this example, and D 6= 0 so a biproper
transfer function will be the result. Examination of the A matrices in (7) and (14) should indicate that the determinants
|sIn A| will be the same for each example. The steps in the process are as before.

s 0 0 7 1 0 s + 7 1 0
sI3 A = 0 s 0 14 0 1 = 14 s 1 (15)
0 0 s 8 0 0 8 0 s
The matrix of minors and matrix of cofactors are
2
s 14s + 8 8s s2 (14s + 8) 8s
M = s s (s + 7) 8
, Cof = s s (s + 7) 2 8
(16)
1 (s + 7) s2 + 7s + 14 1 (s + 7) s + 7s + 14
and the adjoint matrix is

s2 s 1
Adj (sIn A) = (14s + 8) s (s + 7) (s + 7) (17)
8s 8 s2 + 7s + 14
The determinant of (sIn A) is

|sIn A| = (s + 7) s2 + 1 (14s + 8) + 0 (8s) = s3 + 7s2 + 14s + 8 (18)


The steps in the calculation of the numerator are

CAdj (sIn A) = s2 s 1 , CAdj (sIn A) B = 20s2 + 125s + 185 (19)
so the transfer function is
2
20s2 + 125s + 185 5s3 + 55s2 + 195s + 225 5 (s + 3) (s + 5)
H (s) = 3 2
+5= = (20)
s + 7s + 14s + 8 s3 + 7s2 + 14s + 8 (s + 1) (s + 2) (s + 4)

You might also like