Kohonen neural network based approach to voltage weak buses/areas identification

Abstract

The paper presents a neural network based method for the identification of voltage-weak buses/areas. By using power flow analysis and the singular value decomposition method, a Kohonen neural network is trained to cluster/rank buses in terms of voltage stability. Special emphasis is placed on the selection of input information and analysis of the neural network output results. The generalisation capability of the Kohonen network under various operating conditions has been tested on a number of systems. The results obtained for the IEEE 30-bus test system are very encouraging.