Page 173 - Modern Control of DC-Based Power Systems
P. 173

Control Approaches for Parallel Source Converter Systems     137


                 The aim of the controller is to bring the system to the set-points
              values ½x ; 0Š. Therefore, the system variables are modified in order to

                     1
              comply with the Eq. (5.68) as follows:

                                         v 1 5 x 1 2 x
                                                   1                     (5.74)
                                            v 2 5 x 2
                 The nonlinear target system is given by:
                                         _ ξ 52 K 1 ξ 3                  (5.75)


              which has an asymptotically stable equilibrium point in ξ 5 0.

                                                           2
                 The    mapping    function  is   π:R-R ,       meaning    that
              ðv 1 ; v 2 Þ 5 ðπ 1 ξðÞ; π 2 ξðÞÞ. By setting π 1 ξðÞ 5 ξ 5 v 1 , the equilibrium point
              of π 1 ξðÞ 5 ξ 5 0, which is equal to the equilibrium point of v 5 0 and



                                                                      1
              satisfies the hypothesis (5.68).
                 The application of the map π to the system (5.73) in the new variables
              ðv 1 ; v 2 Þ results in a nonlinear system with the partial differential
              Eq. (5.70):
                                         2K 1 ξ 5 ξπ 3
                               8             _
                               >              3     2
                               <
                                    @π 2       2                         (5.76)
                                  2
                                               2
                               >     @ξ  K 1 ξ 5 π 1 ξ 1 c πξðÞð  Þ
                               :
              where u 5 c πξðÞÞ is the control output that drives to zero the off-the-
                        ð
              manifold trajectories and π 1 ξðÞ has been already substituted with ξ. The
              solution of the first equation of (5.76) results in π 2 52 K 1 ξ 2=3 :
                 The implicit definition of the manifold, which forces the trajectories
              of the off-the-manifold coordinate z 5 Φ xðÞ to zero, can be described as:
                           Φ xðÞ 5 v 2 2 π 2 ξ  5 v 2 1 K 1 v 1 2=3  5 0  (5.77)

                                          1 jξ 1 5v 1
                 The dynamic of z must have bounded trajectories to satisfy the theo-
              rem in [28], therefore the time derivative of (5.77) is defined by the fol-
              lowing equation:

                                         _ z 52 K 2 z                    (5.78)

                 The result of (5.78) defines the control output that drives to zero the
              coordinate z and renders the manifold invariant is given by:

                                        2    2=3 3             2=3
                                2
                    u 5 c πðÞ 52 v 2 v 1 2 K 1 v  v 2 K 2 v 2 1 K 1 v    (5.79)
                                2            1  2              1
                                        3
   168   169   170   171   172   173   174   175   176   177   178