Page 808 - The Mechatronics Handbook
P. 808

0066_Frame_C25  Page 7  Wednesday, January 9, 2002  7:05 PM









                       Under mild technical conditions involving controllability and observability of the state space model, the
                       transfer function representations in (25.13) and (25.15) are similar in case the state space model in (25.2)
                       is derived from the differential equation (25.1) and vice versa.


                       25.2 Dynamic Response

                       The Laplace and z-transform offer the possibility to compute the dynamic response of a dynamic system
                       by means of algebraic manipulations. The analysis of the dynamic response gives insight into the dynamic
                       behavior of the system by addressing the response to typical test signals such as impulse, step, and sinusoid
                       excitation of the system.
                         The response can be computed for relatively simple continuous or discrete dynamical systems given
                       by low order differential or difference equations. Both the state space model and the transfer function
                       descriptions provide helpful representations in the analysis of a dynamic system. The result are presented
                       in the following.


                       Pulse and Step Response
                       A possible way to evaluate the response of a dynamic system is by means of pulse and step based test signals.
                       For continuous time systems an input impulse signal is defined as a δ function

                                                                  ∞,  t =  0
                                                  u imp t() := d t() =  
                                                                   0,  t ≠  0


                       with the property
                                                      ∞
                                                     ∫ t=−∞  ft()d t() =  f 0()


                       where f(t) is an integrable function over (−∞, ∞). Although an impulse signal is not practical from an
                       experiment point of view, the computation or simulation of the impulse response gives insight into the
                       transient behavior of the dynamical system.
                         With the properties of the impulse function δ(t) mentioned above, the Laplace transform of the impulse
                       function is given by

                                                             ∞
                                             {
                                            L d t()} =  d s() =  ∫  d t()e – st  t d =  e – s0  =  1
                                                             t=0
                       Hence the output y(s) due to an impulse input is given by y imp (s) = G(s)u imp (s) = G(s)δ(s) = G(s). As a
                       result, an immediate inverse Laplace transform of the continuous time transfer function G(s),

                                                                1
                                                               –
                                                      y imp t() =  L { Gs()}
                       gives the dynamic response y imp (t) of the system to an impulse input response.
                         The computation of the step response is done in a similar way. In (25.9), the Laplace transform of the
                       step signal

                                                               0, t <  0
                                                     u step t()  :=  
                                                               1, t ≥  0
                      ©2002 CRC Press LLC
   803   804   805   806   807   808   809   810   811   812   813