cGANs-based Real-time Stability Region Determination for Inverter-based Systems

In islanded microgrids with droop-controlled inverters, the droop coefficients need to be tuned in real-time to maintain asymptotic stability. Online stability region estimation improves the situational awareness of the system operator and enables the selection of stable droop gains as the system configuration changes. The challenge in the conventional stability region estimation process is its unscalability and high computational complexity. In this paper, an efficient alternative is proposed using conditional Generative Adversarial Networks (cGANs) that demonstrates high accuracy and improved scalability. The implementation of cGANs-based stability assessment is described for an LV distribution test case and its advantages demonstrated.