Power system dynamic security classification using Kohenen neural networks

In this paper a novel approach for transient stability based dynamic security classification and screening of power systems is presented. A Kohenen neural network is implemented as neural network security classifier. The precontingency steady state operating condition of power system is used as the input pattern of the neural network classifier for proper classification of dynamic characteristic. Transient stability is the dynamic behavior by which system security is assessed and classified. Critical clearing time (CCT) is used as security index for both feature extraction and classification of system dynamic security states. For the purpose of feature extraction, correlation between pre contingency operating condition and system dynamic characteristic is used. The proposed approach has been demonstrated on the IEEE-39 bus system with promising results for relatively accurate classification and screening of power system dynamic security.