Page 171 - Modern Control of DC-Based Power Systems
P. 171

Control Approaches for Parallel Source Converter Systems     135


                 The calculation of the control law starts with the definition of the tar-
              get system to map into the nonlinear system to be controlled. The target
              system has typically a low-order dimension, and it is described by the fol-
              lowing expression:
                                          _
                                          ξ 5 αξðÞ                       (5.66)
                                           ξ 5 0                         (5.67)

                 As described in (5.67), the target system has an equilibrium point at
              the origin that must be asymptotically stable.
                 The target system is usually defined a priori by the designer, by speci-
              fying the desired dynamic. This allows the target system to comply with
              the theorem in [28], because it can be defined such as (5.67) is satisfied.
              In the I&I approach, the target system defines also the dynamic of the
              closed-loop system, since the system controlled by the feedback loop con-
                                           _
              tains the copy of the dynamic of ξ 5 αξðÞ.
                                                        p
                                                             n
                 Supposing that there exists a mapping π:R -R such as x 5 πξ ðÞ,
                       p
              with ξAR , the following equations must hold:
                                         πξ ðÞ 5 x                       (5.68)

                                          πð0Þ 5 0                       (5.69)
                 The application of the mapping x 5 πξðÞ and of the control law to the
              nonlinear system (5.65) results in the partial differential Eq. (5.70).
                                                      @π
                               f πξðÞÞ 1 g πξðÞÞc πξðÞÞ 5                (5.70)
                                                      @ξ  αðξÞ
                                              ð
                                ð
                                        ð
                 Once the conditions (5.68) (5.70) are satisfied, the manifold M,
              which is the geometric interpretation of the mapping procedure, can be
              defined implicitly by:

                           n                n jx 5 πξðÞfor some ξAR p    (5.71)
                     f xAR jΦ xðÞ 5 0 5 xARf
                 The Eq. (5.71) shows that the aim of the controller is to drive to zero
              all the off-the-manifold trajectories the manifold, described by z 5 Φ xðÞ.
              The implicit definition of the manifold demonstrates that the only trajec-
              tories permitted are the ones that satisfy Eq. (5.71), which are strictly
              connected to the target system.
                 Moreover, the equation demonstrates an equivalence between the map
                 p
                      n
              π:R -R and the manifold, defining a correlation between the mathe-
              matical and geometrical interpretation of the stabilization procedure.
   166   167   168   169   170   171   172   173   174   175   176