A two-stage trust-tech based methodology for excitation system equivalence

Large-scale power system dynamic analysis usually bears large storage and massive calculation burdens. Equivalence of complex excitation systems is an effective approach to decrease computational complexity with due accuracy. In this paper, we propose a two-stage Trust-Tech based method for deriving excitation system equivalence which aims to provide an optimal equivalent model which satisfies users´ requirements. At the first stage, Trust-Tech methodology is used to find high-quality optimal solutions, if not the global optimal solution. This stage offers multiple local optimal alternatives, which ensures the availability of suitable equivalent models to meet different engineering requirements from users. At the second stage, critical clearing time and H infinity performance indexes are applied to provide useful information to evaluate the alternative sets. Thus, users can select a satisfying solution according to the index value or the practical requirements. Finally, a case is included to verify the effectiveness of the method.