Estimating Rotor Angle and Stability of Synchronous Generator using Neural Network Modeling

Abstract

There are several methods to study the stability and to estimate parameters of synchronous generators in different models [1,2,3]. In these methods, it is assumed that the rotor angle is measurable, while the lack of some signals in transmission line is possible [4]. The main approach of this paper is to estimate rotor angle in a synchronous generator using an artificial neural network (ANN) and dynamic parameters of generator such as electromagnetic torque, mechanical speed, generator current and generator voltage. This way, it is plausible to predict the stability of every generator with a system under error probability. Error has been tested in two ways in this system: first, increased torque load, and second, 3-phase short circuit error. The proposed method has been applied successfully. The simulation results completely confirm the proposed method.