Page 726 - The Mechatronics Handbook
P. 726

0066_Frame_C23  Page 34  Wednesday, January 9, 2002  1:55 PM









                       and provides the frequency domain input–output relation of the system. In particular, the Bode diagram
                       is evaluated as |H(z)|  and arg H(z) for z = exp(iω k T) and for |ω k | < ω n  = π/T, that is, when H(z) is
                       evaluated for frequency points up to the Nyquist frequency ω n  along the unit circle.

                       State-Space Systems

                       Alternatives to the input–output representations by means of transfer functions are the state-space
                       representations. Consider the following finite dimensional discrete state-space equation with a state vector
                       x k ∈    ,  input  u k ∈    ,  and observations  y n ∈    .
                                         p
                            n
                                                              m
                                                 x k+1 =  Φx k +  Γu k
                                                
                                                                 k =  0, 1,…                   (23.65)
                                                
                                                  y k =  Cx k +  Du k


                       with the pulse transfer function


                                                            (
                                                   Hz() =  CzI Φ) Γ +  D                        (23.66)
                                                                   –
                                                                   1
                                                               –
                       and the output variable
                                                         ∞
                                                 Yz() =  C ∑ Φ z x 0 + Hz()Uz()                 (23.67)
                                                             k –
                                                               k
                                                         k=0
                       where possible effects of initial conditions x 0  appear as the first term. Notice that the initial conditions
                       x 0  can be viewed as the net effects of the input in the time interval (−∞, 0).

                       Digital Systems Described by Difference Equations (ARMAX Models)
                       An important class of nonstationary stochastic processes is one in which some deterministic response to
                       an external input and a stationary stochastic process are superimposed. This is relevant, for instance,
                       when the external input cannot be effectively described by some probabilistic distribution. A discrete-
                       time model can be formulated in the form of a difference equation with an external input {u k } that is
                       usually considered to be known:

                               y k =  – a 1 y k−1 – …  – a n y k−n +  b 1 u k−1 +  … + b n u k−n +  w k +  c 1 w k−1 + …  +  c n w k−n  (23.68)


                         Application of the z transform permits formulation of Eq. (23.68) as

                                                             1
                                                             –
                                                           (
                                                                         1
                                              Az (  – 1 )Yz() =  Bz )Uz() +  Cz )Wz()           (23.69)
                                                                      (
                                                                        –
                       where
                                                               –
                                                                1
                                                 Az (  – 1 ) =  1 +  a 1 z +  … +  a n z – n
                                                  Bz (  – 1 ) =  1 +  b 1 z + …  +  b n z –  n  (23.70)
                                                               –
                                                                1
                                                     –
                                                   (
                                                 Cz ) =   1 +  c 1 z + …  +  c n z – n
                                                     1
                                                                1
                                                               –
                      ©2002 CRC Press LLC
   721   722   723   724   725   726   727   728   729   730   731