Small-Signal Modeling and Region-Based Stability Analysis Using Support Vector Machine (SVM) for Autonomous Hybrid AC and DC Microgrids

Hybrid AC and DC MGs provide an effective solution for leveraging distributed energy resources (DERs) to support local AC and DC loads. However, the complex structure and control dynamics of hybrid AC and DC MGs challenge their operational stability, especially when compared to individual AC or DC MGs. This paper develops a holistic small-signal model of autonomous hybrid AC and DC MGs, which incorporates AC and DC sections as well as interface converters between AC and DC buses. The small-signal stability of the developed model is analyzed to derive a stability region. In particular, a 2-D stability region is defined using flexible cross-domain parameter combinations from either control systems or main power circuits of hybrid AC and DC systems. Further, an artificial intelligence (AI)-aided support vector machine (SVM) algorithm is employed to estimate the stability region boundary (SRB), which provides a computationally efficient way to visualize the stable operation region. Leveraging the real-time hardware-in-the-loop (RT-HIL) platform, case studies conducted on a test hybrid system with five converters (two DC-AC inverters, two DC-DC converters, and one interlink converter) demonstrate the effectiveness of the proposed work. Multi-fold cross-validation of all the case studies confirms that the out-of-sample misclassification rate meets the required threshold of 5%.