Features selection for training generator excitation neurocontroller using statistical methods

Abstract

Essentially, control system requires suitable control signal for yielding desired response of a physical process.Control of synchronous generator has always remained very critical in power system operation and control. For certain well known reasons power generators are normally operated well below their steady state stability limit. This raises demand for efficient and fast controllers. Artificial intelligence has been reported to give revolutionary outcomes in the field of control engineering. The capability of Artificial Neural Network (ANN) to map any nonlinear function satisfactorily based on input-output data has been widely established in intelligent control. Selecting optimum features to train a neurocontroller is very critical because correlation between features of parameters may avert learning capability of an ANN. In this work statistical methods are employed to select independent factors for ANN training. © 2011 Springer-Verlag.