Active Power Oscillation Property Classification of Electric Power Systems Based on SVM

• Main Authors: Ju Liu, Wei Yao, Jinyu Wen, Haibo He, Xueyang Zheng
• Format: Article
• Language: English
• Published: Hindawi Limited 2014-01-01
• Series: Journal of Applied Mathematics
• Online Access: http://dx.doi.org/10.1155/2014/218647

Nowadays, low frequency oscillation has become a major problem threatening the security of large-scale interconnected power systems. According to generation mechanism, active power oscillation of electric power systems can be classified into two categories: free oscillation and forced oscillation. The former results from poor or negative damping ratio of power system and external periodic disturbance may lead to the latter. Thus control strategies to suppress the oscillations are totally different. Distinction from each other of those two different kinds of power oscillations becomes a precondition for suppressing the oscillations with proper measures. This paper proposes a practical approach for power oscillation classification by identifying real-time power oscillation curves. Hilbert transform is employed to obtain envelope curves of the power oscillation curves. Twenty sampling points of the envelope curve are selected as the feature matrices to train and test the supporting vector machine (SVM). The tests on the 16-machine 68-bus benchmark power system and a real power system in China indicate that the proposed oscillation classification method is of high precision.