Dynamic state estimation in power system using a gain-scheduled nonlinear observer

Time-optimal control of network impedances by a series capacitor has been considered to be a powerful method to enhance transient stability in power systems. A time-optimal control scheme relies on the availability of the synchronous machine angle and speed variables. However, a series capacitor will, in general, not be located near any synchronous machines, such that angle and speed variables cannot be measured locally. In this paper the authors propose a gain-scheduled nonlinear observer to provide estimates of the system states for the control design. Kalman filter design and gain scheduling techniques are used in this procedure. The method is verified by the time simulation of a single-machine infinite-bus power system and is shown to give satisfactory results