Support Vector Classifier with enhanced feature selection for transient stability evaluation

Abstract

Today's power transmission systems have a tendency to operate closer and closer to their stability limits. In this scenario, there have been continuous efforts to develop new techniques and tools for assessing the stability status of power systems. This paper presents a Support Vector Classifier (SVC) to identify the transient stability of power systems subjected to severe disturbances. The nonlinear relationship between the pre-fault, during-fault and post-fault power system parameters and the stability status of the system under post-fault state is captured by the SVC trained offline. Significant generators are selected by feature selection based on the sensitivity of stability margin and the features other than generators are selected based on a step wise feature selection by three fold cross validation. The performance of the proposed SVC is demonstrated through the simulations carried out on the IEEE 17 generator reduced Iowa system.

Keywords

• transient stability
• Support Vector Classifier (SVC)
• dimensionality reduction
• feature selection
• energy margin