Optimal Power System Stabilization Through Excitation and/or Governor Control

Abstract

For an optimal linear regulator design a performance function of the quadratic form must be chosen. The question arises of how to decide the weighting matrix Q of the performance function. A new method is developed in this paper to determine Q in conjunction with a left shift of the dominant eigenvalues as far as the practical controllers permit. The method is then applied to the optimal control design of a typical power system. Three cases are investigated, the first with an optimal excitation control uE, the second with optimal governor controls uG and uG, with and without the dash-pot, and the third with uE plus uG control. The stabilizing signals thus obtained are given nonlinear tests on the same power system. It is found from the results that the optimal controls are more effective than conventional excitation control, that the optimal governor control without dash-pot is just as good as the optimal excitation control, and that the optimal uE plus uG control is the best way to stabilize a power system.