A preventive transient stability control method based on support vector machine

Abstract This paper presents a support vector machine (SVM)-based method for preventive transient stability control (TSC) of power systems. The features of the SVM model used in TSC are the same as in transient stability assessment (TSA). The features include non-control variables so as to improve the accuracy of TSA and sensitivities. A hybrid method combining SVM and time-domain simulation is also proposed to increase the TSA accuracy. First, this paper defines a transient stability assessment index based on SVM model. Second, the sensitivities of the SVM-based transient stability assessment index with respect to control variables (generator active power) are calculated and ranked to select control generators from them. Then, the power shifting amount of all the control generators and balance generators is calculated and the corresponding control scheme is verified by the hybrid TSA method. If generation redispatching measures are invalid, then load shedding measures are used at the same time. Finally, comprehensive studies are conducted on the WECC 9-bus system and a certain provincial 2036-bus system in China. The results reveal that the proposed hybrid TSA method increases the classification accuracy rates from 99.0%, 99.3% to 100% for the certain provincial 2036-bus system, and effective control measures are obtained with the proposed SVM-based TSC method for both systems.