This thesis studies the modeling and analysis of power System containing static VAR compensators (SVC). Low-order models, calculated by Hankel singular-value truncation réduction and N4Sid dynamic identification, provide the foundation for the modeling and analysis techniques. For cases where many SVCs or FACTS (Flexible AC Transmission Systems) devices are installed in a power System, a new subsystem structure based on thèse low-order modeling techniques has been developed. The proposed low-order models allow the computation of low-order LQG-type (Linear Quadratic Gaussian Regulator) control laws, as well as low-order robust controls, which provide a better performance relative to traditional SVC controls (proportional type). Using the subsystem structure formulated in this thesis, methods for calculating decentralized controllers for power Systems hâve been conceived. In addition, through the use of structured singular values, a new robustness analysis technique has been developed which allows the détermination of stability boundaries through the calculation of closed-loop transfer functions. The theoretical approaches in this work hâve been applied to a test power system containing 11 nodes, as well as a larger 173-node test system.
