Kooperationspartner:
Lehrstuhl für Regelungstechnik, Universität der Bundeswehr, Hamburg
Prof. Dr.-Ing. Joachim Horn, Dipl.-Ing. Ingo Winzenick
The dynamical behavior of power systems is determined by the load characteristics, mainly. Therefore, in the past a lot of
load models are constructed and they are verified much more simulational studies, which is reported in literature very
well. The problem with these studies is, that most of them are based on time series analysis only. From literature it is also
known, that newton based algorithms for simulation can show a fractal structure. All calculated solutions are therefore
singular. Moreover, it could be shown, that the interconnection of nonlinear models of power systems devices to more
complex power systems lead to a numerial index problem. Even in the simplest case of a power system representation by
load flow equations, there exist a lot of solutions, only some of them make sense in a physical manner, but numerically they
exist. On the other hand, there is a need in much more detailed information about power system dynamics. The introduc-tion
of FACTS–devices to force the system to have special dynamical characteristics under various conditions makes the
problem not simpler.
One possibility to overcome some of the problems is, to apply continuation methods, to calculate all possible solutions of a
power system under study. The application of continiuation methods on power system analysis and computation of dy-namics
has the advantage, that beside a possible calculation of time series, the complete solution of the system in depen-dence
of interesting parameters can be achived. Furthermore, sophisticated software programs can classify the properties
which every calculated solution point has. Special bifurcational properties can be continued.
From the analysis and numerical computations, an outlook will be given to use the results to determine FACTS–devices to
manipulate the bifurcational behavior of the systems under consideration.
