Research on UPFC Fault Diagnosis Based on KFCM and Support Vector Machine

In order to make full use of limited data to protect the normal and stable operation of Unified Power Flow Controller (UPFC) prospectively, a fault diagnosis method based on KFCM and Support Vector Machine (SVM) is proposed in the study. Above all, the DC voltage signal is used as the monitoring signal to determine seven kind of characteristic quantity corresponding to the simulated fault types to obtain the initial data sources. Then, the data are sent to KFCM algorithm for clustering to verify the validity of the feature vectors and the number of classifications. Finally, the feature data are sent to SVM to further verify the classification results. In conclusion, the validity and accuracy of the method were proved by simulation experiments and shown that the UPFC fault mentioned in this paper can be recognized through the proposed method.