An analysis of relative performance of state variables in the design of power system stabilizer through neural networks

In this paper, a new approach to the design of power system stabilizers, which is a control element, that increases the stability of generators against low level frequency oscillations is investigated and analyzed by the use of neural networks. It is also shown that the determination of state variables, while designing power system stabilizers with artificial neural network (ANN-PSS), takes a very important place. This is determined by investigating the relative performance of state variables used in the design of ANN-PSS. The aim of this paper is to seek a strong correlation among the state variables that will give the best results for the PSS-ANN design. Therefore, seven different types of power system stabilizers with ANN (ANN-PSS), have been proposed.