Enforcing coherency in droop-controlled inverter networks through use of advanced voltage regulation and virtual impedance

There is an increasing need to model large networks of inverter-based distributed generation due to the limits of existing simulation tools. A deliberate enforcement of intra-feeder coherency between adjacent droop-controlled inverters is a worthwhile objective, as phase angle coherency will allow for a deliberate partitioning of fast and slow dynamics within the network. This partitioning, in turn, results in simpler and more tractable reduced-order network dynamics that are easier to model. This paper presents the adaptation of the generalized eigenvalue perturbation approach to the identification of coherent groups within grid-forming, droop-controlled inverter networks. For many networks, a rigorously-derived condition of coherency can be difficult to achieve, given the expected range of L-C-L filter and grid-tie impedances. To remedy this limitation, this paper investigates the use of multi-loop voltage regulation and virtual impedance to enforce the conditions that allow for coherency, and thereby more rigorously justify the use of simple nonlinear aggregate inverter models to accurately model large sections of the inverter network.