A new methodology for the design of power system stabilizers

In the wake of energy crises and resource crunches, power systems today are called to operate under highly stressed conditions. Under such conditions, the ability of the system to recover from disturbances assumes paramount importance. A potentially unstable system can endanger system security and limit power transfers which have economic implications. From a control view point, the state of the art primary controllers for power systems are based on output feedback which cannot meet high performance objectives. On the other hand, complete state feedback and nonlinear controllers can guarantee high performance objectives. An accurate knowledge of the dynamic state of the synchronous machine is useful for designing sophisticated excitation and supplementary control systems. The question is whether such controls can be effectively designed and implemented. A fundamental impediment to the design of such controls has been the lack of accurate information on the dynamic state of the synchronous generator.; In this thesis, a systematic and analytic framework is developed to compute the dynamic state of a synchronous generator connected to an arbitrary external network. The method employs locally available measurements at the machine terminal and the fundamental electrical parameters of the machine to arrive at an estimate of the dynamic state. The strength of the proposed method lies in the fact that no knowledge of the network parameters or its associated topology is required. Based on this framework, a control architecture for primary control design is proposed. In this scheme, an accurate estimate of the dynamic machine state together with the output are employed to design the excitation control. As an application example, a nonlinear excitation control scheme is designed based on the small signal criterion after the cancellation of certain nonlinear terms. Numerical simulations indicate that the proposed control significantly enhances transient stability even under highly stressed system conditions and is robust enough to variations in operating conditions. Broadly, this is indicative of the potential benefits of nonlinear control. It is anticipated that the framework proposed in this thesis would open the roads to the development and design of several advanced nonlinear control strategies for power systems control.