Stability Monitoring and Controller Autotuning of Power Converters in DC Microgrid

Recently, there has been an increasing interest towards grouping several power resources together with some loads as well as some energy storage systems in a microgrid environment. This is mainly because a high number of distributed energy resources (DERs), such as renewable energies and energy storage systems can be integrated in a microgrid environment, that, in turn, will lead to a reduction in the transmission and distribution losses, the overall system costs, as well as the CO2 emissions. In addition, as the generation is going to be mostly near to the consumption point, the power quality, eciency and reliability will be signicantly increased. Microgrids are also a smart choice for the remote locations that are beyond reach of the current grid. Dc microgrids bring with some advantages over their ac counter part. For instance, they are more compatible with the dc nature of many DERs such as photovoltaics and energy storage systems. Also, the inductive voltage drop is removed in a dc system. Thus, a large number of DERs can be integrated into a dc microgrid by taking advantage of power electronic converters, that introduce several control and operation benets. Power converters used in dc microgrids are usually equipped with several control loops. When many converters are connected to a common dc bus, the dynamic performance of some control loops may be dierent from the behavior designed for the stand-alone converter, due to possible eects of the interconnected converters. This issuewhich is typically referred to as the `interaction effect' of multiple parallel converters can lead to stability and performance concerns in a dc microgrid. Thus, interaction eect on a generic control loop depends on the interconnected power converters, for instance, on their topology, control structure, parameters, etc. In order to know the real-time control performance and stability of the control loops within dc microgrid power converters, it is important to equip the converters with online stability monitoring tools. The monitored data will not only include the internal stability conditions of each loop, but also take the interaction eect into account. Subsequently, some corrective actions can be introduced in the system to maintain a desired dynamic performance and avoid instability. In addition, in the context of smart microgrids, the advanced monitoring tools, as well as adaptive control and management actions are of a wide interest. This work rstly, investigates an on-line stability monitoring technique that is inspired by the Middlebrook's injection method. This method allows to estimate and monitor the stability margins of a generic control loop (e.g., current loop, voltage loop, droop loops, etc.) within dc microgrid power converters. Since we target a multi-converter environment, the presence of multiple perturbations coming from the monitoring units of several converters is also taken into account. Secondly, two dierent on-line tuning techniques are proposed, that both aim to achieve the desired phase margin for a generic control loop at the reference bandwidth. These methods are based on injecting a small-signal perturbation at the desired reference crossover frequency into the loop under study. In other caseswhere a full picture about the performance of dierent loops over the entire bandwidth is desiredmultiple orthogonal pseudo-random binary sequences (PRBSs) are proposed to be simultaneously injected in several control loops. This will provide the frequency responses of all the loops in a single measurement cycle. Finally, in order to further assess the microgrid-level stability and dynamic performance, some of the monitored data are eectively used to estimate the dc bus impedance, which has been shown to provide a measure of the stability and performance of the entire microgrid. All the stability monitoring and adaptive tuning functions are experimentally validated in a laboratory setup that emulates a dc microgrid