ARTIFICIAL INTELLIGENCE APPROACH TO STABILITY OF POWER SYSTEMS

Abstract

The Critical Clearing Time (CCT) of dynamic/transient stability analysis represents maximum fault curation beyond which the system loses synchronism. It is used as a figure of merit to compare results obtained by time simulation with those obtained by direct methods. The CCT is a complex

function of the pre-fault system conditions

(operating

point, topology, system parameters), fault structure ( type and location), and post-fault conditions that are dependent on the protective relaying strategy.

This thesis presents an approach based on Artificial Neural

Network (ANN) for transient stability assessment of power

system. The neural network has two processes; learning and classification. In the learning process. the network is

presented with a pair of patterns, an inpit pattern and a corresponding target output pattern. In tle classification process, the user provide a pattern of system description parameters to the neural network and the network returns an estimate of the output patterns.