Reactive power dispatch incooperating voltage stability improvement using artificial neural network / Rasida Norjali

This thesis presents the development of an Artificial Neural Network (ANN) based technique for reactive power dispatch that aims to improve voltage stability of a power system. In this study, a multi-layer feed forward ANN with error back propagation algorithm was used. The proposed method was tested on two models, which are the 6 bus, and IEEE 14 bus interconnected systems. The testing and training data were generated by Fast Decoupled Load Flow method and the voltage stability at a load bus was measured by evaluating the voltage stability index, i.e. L-factor, developed in reference [1]. The results show that the ANN could be used to determine the value of reactive power and to predict voltage stability level for power system, since they are in close agreement with the calculating results.