Robust controller for damping power system oscillations

Design of power system stabilizer using Linear Quadratic Regulator (LQR) technique. Designs that ignore noise in a plant are likely to fail when implemented in actual conditions. Presence of the noise, which results in LQR output oscillating with very small magnitude. This type of unwanted noisy output is eliminated using Linear Quadratic Gaussian (LQG) Regulator. This paper presents techniques for the design of power system stabilizer using LQG regulator comprises the Kalman estimator, which is obtained from the optimal state feedback gain designed with LQR. The LQG regulator minimizes the some quadratic cost function that trades off regulation performance and control effort. This indicates that the LQG compensated system is more robust with respect to noise than the LQR with same regulator gain matrix.