Main Article Content
This paper deals with the applying of the Linear Quadratic Gaussian (LQG) technique to the design of the robust controller for two models of power system. The first model represents only the electrical control part of power system by means synchronous generator connected to infinite bus, while the second model, adding a turbine and governor to the model 1. Combined the Kalman filter, which is an optimal observer with the optimal LQR regulator to construct the optimal LQG controller are evaluated. Electromechanical oscillations of small magnitude and low frequency exist in the power system operation and often persist for long periods of time. Power system stabilizers (PSSs) are traditionally used to provide damping torque for the synchronous generators to suppress the oscillations by generating supplementary control signals for the generator excitation system. Numerous techniques have previously been proposed to design PSSs but many of them are synthesized based on a linearized model. The dynamic characteristics of the proposed PSS are studied in a typical synchronous machine connected to infinite-bus of power system through transmission line. Simulation results show the proposed PSS is robust for such nonlinear dynamic system and achieves better performance than the conventional PSS in damping oscillations.