Automatic reactive power and voltage control for regional power grid based on SVM

In this paper, the traditional reactive power and voltage control problem is regarded as a multi-class classification problem, and a novel approach based on Support Vector Machine (SVM) classifier is proposed. According to the approach, the power grid operating status is classified according to the power factor and voltage at each substation and the corresponding control strategy is selected to control the capacitors and transformer taps. A naive progressive learning strategy is also presented to make sure the classifier can keep learning in the operation process. The approach is suitable for online operation. The decision results are robust and coordination operation between the substations can be achieved. A simple radial system containing three substations is used for case study. The results illustrate the effectiveness of the proposed approach.