DESIGN OF ROBUST POWER SYSTEM CONTROLLERS

Abstract

Power system is such a complex structure that it is not always easy to derive an exact model for it. These complexities in modeling leads to model errors which are also called as `model uncertainties'. If a controller is designed for a particular control problem in the power system, without considering these model uncertainties, that controller may fail in real application. So a design methodology which can take these model uncertainties into account is necessary specifically to the power system control problems. Structured singular value (μ) synthesis is the design method which is having the ability to deal with these model uncertainties. So a controller that is designed using this method will be robust enough to work on the real power system. In this dissertation work, a deregulated power system load frequency control problem is considered for explaining the development of the μ-synthesis from the H- synthesis. Three types of controllers are designed for the deregulated power system load frequency control problem, namely • Hco- controller • Weighted H.-controller 0 g-controller `Bounded complex 'uncertainty' models are developed for the Damping coefficient 'uncertainties in the deregulated power system model. Along with these two uncertainties, the neglected high frequency dynamics uncertainty due to the first order approximation of turbines and governors is also considered. A ti-controller is designed after taking these model uncertainties into account. The robustness properties of this μ-controller are compared with the weighted H- controller with the help of time response simulations. Controller order reduction technique is applied on this ti-controller to reduce it to a 3`d order controller from 25th order. The robustness of this third order controller is also checked using time response simulations. For this work, the various algorithms available in the Robust Control Toolbox of MATLAB® are used. g-controller is designed using the DKITGUI tool available in this Robust Control Toolbox of MATLAB®.