Fast dynamic stability analysis of a power system using artificial neural networks

Abstract

This paper deals with the methodology of analysing the dynamic stability of a power system, based on the application of a variety of architecture and configurations of artificial neural networks. On the basis of the complete system state vector, which represents the input into a multi-layered neural network, the topology detection of the transmission network is carried out. Depending on the operating regime - the system topology and load level - the corresponding multi-layered neural network is activated in order to detect the stability of the system. Stability analysis is achieved by tracking changes in the dominant eigenvalues of the system which are the output of the trained neural network. Finally, the paper presents a means of analysing the dynamic stability of the power system on the basis of the reduced state vector, obtained by extraction of the characteristic state coordinates by means of self-organizing neural network.