Olaoluwapo Ajala with A. Domínguez-García
The popularity of the microgrid has grown as adoption of distributed energy resources (DERS) has accelerated. Loosely speaking, a microgrid is defined as a collection of DERs and loads interconnected via an electrical network with a small physical footprint, which can operate as part of a large power system (grid-connected mode), or as an autonomous power system (islanded mode). A key advantage of microgrids present is that they can operate as a single controllable entity. As a result, the net effect of a microgrid’s DERs and loads can be monitored and controlled to maintain operational reliability of the power system.
When microgrids in a power system increase, a need to seamlessly them arises. This requires a synchronization of the microgrid’s interconnection points, i.e., ensuring that their voltage magnitude, frequency, phase, and phase sequence matches. In order to achieve synchronization of multiple microgrids, the outputs of each DER in the microgrids must be strategically adjusted — not usually a trivial task.
Our main contribution is the development of a method for synchronizing two or more microgrids operating in islanded mode, with the goal of electrically interconnecting them; the proposed method is robust to disturbances or errors in the information exchanged between the microgrids that are trying to synchronize. Our approach to microgrid synchronization is based on casting the synchronization problem as an observer design problem. We have preliminary results that involve synchronizing two microgrids, each having a single electrical power generator, and we are working to extend these results for use in a more complex system.
In Figures 2 and 3, the synchronization error, i.e., the frequency difference of the interconnection points, and the bus frequency are depicted for different measurement disturbance values d(t). We assume that the voltage magnitude and phase sequence of the interconnection points match. In the depicted results, the two microgrids are interconnected only when (i) the observed phase difference of the connection points is a multiple of 2π and (ii) the synchronization error is within admissible bounds specified by IEEE standards; these bounds are depicted using the color red. Examining the results in Figure 2, we observed that for: (i) d(t) = 0.125 π rad, the microgrid connection points successfully synchronized around t = 200 s, (ii) d(t) = 0.25π rad, the microgrid connection points successfully synchronized around t = 260 s, and (iii) d(t) = 0.5π rad, the microgrid connection points fail to synchronize. This research is supported by the Department of Energy.