Eigenvalue-Based Stability Analysis for Droop-Free Controlled Islanded Microgrid with Symmetric/Asymmetric Communication Network

Applying droop-based hierarchical control for islanded microgrids would face several unavoidable operation issues, such as large frequency deviation, high dynamic fluctuation, and inefficient coordination between different control layers. To this end, droop-free control has been explored to achieve the power sharing goal through neighboring communication, while maintaining system frequency levels (i.e., keeping the average nodal frequency constant). Although droop-free control shows potential in mitigating the defects of droop-based control, current research on the theoretical understanding of its stability performance remains limited. Hereby, this paper discusses the stability performance of droop-free controlled islanded microgrid via eigenvalue-based analysis. Specifically, a modified eigenvalue analysis method is first proposed to prove that all effective system eigenvalues under the symmetric communication network are located in the left half-plane, indicating that the symmetric droop-free controlled microgrid is asymptotically stable. Then, the asymmetric communication network is discussed, and the analytical eigenvalues with symmetric/asymmetric communication network designs are deduced under the homogenized electrical network. With the deduced analytical eigenvalues, stability performances of the symmetric/asymmetric designs are assessed via stability margin analysis and vulnerability analysis. Numerical case studies illustrate stability performance of the proposed designs, demonstrating that the asymmetric cycle design outperforms symmetric cycle and symmetric/asymmetric chain designs in overall convergence speed and stability against communication errors between droop-free controllers.