No abstract

Abstract-A dc nanogrid is a hybrid renewable system since renewable sources supply the average load demand, while storage and nonrenewable generation maintain the power balance in the presence of the stochastic renewable sources. The system is power electronic based, with converters being used to interface both the sources and loads to the system. The nanogrid is controlled using dc-bus signaling (DBS), a distributed control strategy in which the control nodes, the source/storage interface converters, induce voltage-level changes to communicate with the other control nodes. This paper explains the control structure required for the converters to permit the use of DBS, and explains a procedure for implementing a system-wide control law through independent control of the source/storage interface converters. Experimental results are presented to demonstrate the operation of this novel control strategy.

As the penetration of complex grid connected devices, such as power electronic inverters, increases, so too does the complexity of analysing system stability. A graphical application of the Nyquist Stability Criterion is presented that indicates how an individual load and source each contribute to the closed-loop system grid eigen values. The case study is not limited to particular impedance forms or scenarios like the more common complex torque coefficients or passivity theory methods. From the individual frequency responses of the load and source impedances the graphical technique indicates how each impedance contributes to the system stability. Examples are provided that successfully indicate the cause of instability for a digitally controlled Voltage Source Inverter (VSI) operating as a microgrid, with a Current Source Inverter as a load. A second example is provided that identifies potential instability of a VSI running an induction machine. A 42kW inverter system is used to confirm the findings, showing a close correlation with the theoretical analysis.

Abstract-The increasing use of grid-connected inverter systems is resulting in a desire for parallel-connected inverters that offer greater power capacity while maintaining the high control bandwidth achieved by individual inverters. This paper demonstrates that, in addition to the traditional stability and bandwidth limitations of digitally controlled inverters, further stability and bandwidth limitations occur when LCL inverters with a common set point are connected in parallel to a grid. This paper provides detailed discrete-time derivations for parallel grid-connected inverters and uncovers stability and bandwidth limitations that only occur in grid-connected applications and are not apparent if the system is studied in continuous time. This paper demonstrates that, in a typical application, the voltage bandwidth of an LCL parallel inverter array is 25% lower than a single module or LC parallel configuration. Both simulations and hardware demonstrations on a 105-kVA parallel three-module grid-connected system confirm the findings.