Abstract: This study presents a multi-class Support Vector Machine (SVM) based method for on-line static security assessment of power systems. To classify the system security status, a group of binary SVMs have been trained. The multi-class Fisher score has been used for feature selection algorithm and the data selection has been done based on the confidence measure, to reduce the problem size and consequently to reduce the training time. The proposed method has been applied to New England 39-bus power system. The simulation results demonstrate the effectiveness and stability of the proposed method for on-line static security assessment procedure of large scale power systems.