Classification of power system disturbances based on wide-area frequency measurements

The automatic disturbance detection and classification based on wide area frequency data from a nationwide frequency monitoring network (FNET) is very important to improve control algorithm for ensuring power system security and reliability, an essential function for smart grid infrastructure. Moreover, in a future power system mostly consisting of distributed generators and renewable energy resources on which the disturbance has more impact, the analysis of disturbances by classifying and categorizing realtime frequency data is rather critical. Many strategies are used right now in on-line analysis, but none of them has good performance. The big background noise and the irregular pattern characters of each kind of disturbances are main reasons. Artificial intelligence method could be one of the solutions. But the lacking of enough training examples has been the bottleneck to apply the artificial intelligence method in practice. Therefore, two methods to classify power system common disturbances is presented in this paper: an artificial neural network (ANN) and support vector machine (SVM) are designed to discern the otherwise hard-to-classify disturbances pattern, and by constructing the disturbances mathematical model, enough examples are obtained conveniently to train the ANN and yield good performance. In the end of the paper, verification of the constructed ANN and SVMs performance is given by using real frequency data from FNET.