Model predictive control of facts devices for power system transient stability

Abstract

Drawing on the recent availability of, and advances in, high-speed flexible AC transmission systems (FACTS) devices, wide-area measurements (WAMs) and high-performance computer systems, the present paper has the objective of developing a real-time control methodology for enhancing and/or maintaining power system transient stability. The detailed dynamic model of the power system in the prevailing operating condition is first linearized, taking into account the system responses following disturbances. The time-varying linearized power system model forms the basis for model predictive control (MPC) adopted for developing the methodology. The control is subdivided into a number of time horizons in the period relevant to the transient stability time frame. The system dynamic model is used in a predictive mode in each time horizon. At the start of each time horizon, for which the current system dynamic response is available via WAMs, subsequent system responses within the time horizon are predicted using the system dynamic model, and optimized, subject to FACTS devices operating limits, to obtain maximum system damping with respect to transient stability. The variables in the optimization are the FACTS devices input references. Their optimal values obtained from the optimization are used for setting the input references to the FACTS devices controllers. The control is applied repeatedly for successive time horizons in each of which an optimal set of FACTS devices input references is derived and implemented. The effectiveness of the control methodology developed in the paper is illustrated by dynamic simulation using a test power system having a FACTS device of the thyristor-controller series capacitor (TCSC) type.