# ELG4152DCH13.

doc

Chapter 13: Digital control systems 1. AD /DA converter

continues signal r (t ) = sin(2π ft ) , discrete signal r (nT ) = sin(2πfnT ) , n = 1,2,3... , a sequence of values occurred at time nT. The output signal of an ideal sampler is r * (t ) = r (nT )δ (t − nT ) , δ (t − nT ) is the impulse happened at time instant t − nT = 0 ,i.e. t = nT

2. s transform and z transform (Chapter 2 and Chapter 13): to linearize the nonlinear operation, for example, in s domain serial connection

F (s) =

0−

f (t )e − st dt
σ + j∞

define z = e sT , integral to summation

X ( z) =

n = −∞

∑ x ( n) z

−n

f (t ) =

1 F ( s )e st ds ∫ 2πj σ − j∞

x ( n) =

1 X ( z ) z n −1 dz ∫ c 2πj

Table 13.1 summary of s and z transform

3. operations in z domain transfer function in z domain is defined as G ( z ) = Y ( z) ; R( z ) Y ( z) G( z)H ( z) = and the closed loop feedback system is T ( z ) = R( z ) 1 ± G ( z ) H ( z )

4. stability analysis in the z-plane z = e sT = e (σ + jω )T = e σT e jωT = e σT (cos(ωT ) + j sin(ωT )) = eσT ∠ωT

1

whatever the system transfer function is.3 G ( s ) = G0 ( s )G p ( s ) = k k k 1 − e − sT 1 = (1 − e − sT )( 1 + 2 + 3 ) 2 s s ( s + 1) s s +1 s G ( z ) = (1 − z −1 ) Ζ(k1t + k 2 u (t ) + k 3 e − t ) . x(n) = u (nT ) . 5.doc In the left-hand s-plane. + k2 + k3 2 z −1 ( z + 1) z − e −T Where k1 = 1 1 | s =0 = 1 s 2 | s =0 = s ⋅ s (s + 1) s +1 ∂ −1 1 | s =0 = −1 k2 = ( s 2 ) | s =0 = ∂s s ⋅ s (s + 1) ( s + 1) 2 1 1 ( s + 1) | s = −1 = 2 | s = −1 = 1 k3 = s ⋅ s (s + 1) s 6. x(n) = e − anT .ELG4152DCH13. x(n) = δ (nT ) . The root locus of digital systems (Chapter 7 and Section 13. σ < 0 . X ( z ) = ∑ x(n) z − n = ∑ u (nT ) z − n = ∑1z − n = −∞ −∞ 0 ∞ ∞ ∞ 1 1 − z −1 x(t ) = e − at . X ( z ) = ∑ x(n) z − n = ∑ e − anT z − n = −∞ −∞ ∞ ∞ 1 1− e − aT z −1 example 13. X ( z ) = ∑ x(n) z − n = ∑ δ (nT ) z − n = δ (0) z −0 + 0 = 1 x(t ) = δ (t − kT ) .10) Y ( s) p( s) = For a system with the transfer function T ( s ) = R( s) q(s) The roots of the characteristic equation q(s) determine the modes of the response of the system. by checking the transform summarized in Table 13. So the characteristic equation should be 2 . First. therefore the related magnitude of z varies between 0 and 1. examples: x(t ) = δ (t ) . The root locus is the path of the roots of the characteristic equation traced out in the s-plane as a system parameter k is changed. x(n) = δ (nT − kT ) .1. So the discrete system is stable if all the poles of the closed-loop transfer function T(z) lie within the unit circle of the z-plane. ∞ ∞ −∞ −∞ −∞ −∞ ∞ ∞ X ( z ) = ∑ x(n) z − n = ∑ δ (nT − kT ) z − n = δ (kT − kT ) z − k + 0 = z − k x(t ) = u (t ) . we get Tz z z G ( z ) = (1 − z −1 )(k1 ). the root locus traced out the root path with respect to only one parameter we are interested.

Example 12.9. For example.7.3333 Any point on the root locus is a root of the characteristic equation corresponding to a specified parameter k. τ1 = −0.13. the root starts from s = − For example. k= ∞ . root is s = − . and 13. 3 .doc rearrange into the form ∆( s ) = q( s ) = 1 + kP( s ) . Table 7. The polynomial 1 + kP( s ) P(s) is called root locus function. Also in Chapter 13. we need to assume the relation among the parameters according to the given information and only concern the design on one parameter. we need to use the specification as sys=P(s). the first : k sτ 1 + 1 sτ 1 + 1 + k 1+ sτ 1 + 1 1 2 k=0. In Chapter 7. the 12-step method of s-plane root locus is explained. k=0. where k is the parameter we are kP( s ) interested. 13.k3…. the similar method is introduced for z-plane root locus. root locus plots for typical transfer functions are summarized. root is s = − 11 τ1 .11. And the corresponding transfer function is T ( s ) = . root is s = − τ1 τ1 k=10. like k1.k2. k=1. with τ 1 = 3 . k sτ 1 + 1 k k T (s) = = . rlocus(sys). root is s = − 1 τ1 to s = −∞ . In chapter 7. If the system T(s) includes several parameters.ELG4152DCH13. When we use Matlab to plot the root locus of system T(s). root is s = −∞ 1 when k increases from 0 to infinity.