Data-analytics for power system stability assessment

Abstract

With the increasing integration of phasor measurement units (PMUs) and supervisory control and data acquisition (SCADA) in power systems, intelligent data-analytics for short-term voltage stability assessment becomes achievable. This task requires fast response and accurate conclusion, especially, to avoid wrong conclusions for the actual unstable cases. Given this, an intelligent post-fault short-term voltage stability (STVS) assessment method is proposed in this research. By introducing Gramian Angular Field (GAF) transform, two-dimensional convolutional neural network (2D-CNN), and adaptive confidence interval (ACI), the proposed method shows better performance to carry out the task. The related tests are based on the New England 10-machine 39-bus system with an obtained 6536-case dataset.