Generator coherency identification using support vector machine.

Generator coherency identification is an important step that needs to be taken in the case of a fault to enable durable islanding that could prevent a cascading blackout. Coherent generators are generally identified using the behaviour of rotor angles of the generators. Machine learning techniques have been implemented to tackle this problem. However, these techniques can be configured in different variations This research aims to implement a fine-tuned machine learning technique into the problem of generator coherency identification, to determine coherent generator groups with extreme accuracy. For this purpose, SVM was chosen as the machine learning technique to be implemented in this research, and was fine-tuned to yield the most accurate results for this problem. Data was collected in the form of rotor angles using a simulation of the IEEE 39-bus test power system in DIgSILENT PowerFactory. Based on the results obtained, SVM proved to be accurate in the primary performance metric RMSE, with a value of 0.077. Furthermore, in the secondary performance metric, R2, the value was extremely high at 0.98. Moreover, in various tests such as with reduced input data of 1000, 500, 250 and 50 out of the original 2000, and reduced computational power of 50% and 90%, SVM was accurate, in both RMSE and R2. A third performance metric, accuracy as a percentage, was also analysed, where SVM had an accuracy of 65.7%, which was compared to an RBFNN model in the literature that had a 95.03% accuracy. However, as other performance metrics show that SVM is accurate, it can be concluded that the fine-tuned SVM is applicable for the clustering of coherent generators. [ABSTRACT FROM AUTHOR]