Key points are not available for this paper at this time.
Whenever control laws are computed on the basis of linear optimal control theory and implemented on non-linear systems, such as power systems, the performance is not as good as expected because of saturation type non-linearities in the system components. The only way to ensure that a control law is adequate, short of actually testing the real system, is to observe the simulated behavior of the system. Departure between a calculated linear control law and the actual response can be minimized by (a) computing a feedback control law with gains of small magnitudes to achieve a pre-assigned set of eigenvalues for the closed loop system and by (b) judicious assignment of the eigenvalues to be achieved. This paper discusses a method for achieving such a control law.
Kumar et al. (Tue,) studied this question.