Return to search

Improved control strategies for droop-controlled inverter-based microgrid

The main focus of this PhD thesis is fundamental investigations into control techniques of inverter-based microgrids. It aims to develop new and improved control techniques to enhance performance and reliability. It focuses on the modelling, stability analysis and control design of parallel inverters in a microgrid. In inverter-based microgrids, the paralleled inverters need to work in both grid-connected mode and stand-alone mode and should be able to transfer seamlessly between the two modes. In grid-connected mode, the inverters control the amount of power injected into the grid. In stand-alone mode, however, the inverters control the island voltage while the output power is dictated by the load. This can be achieved using droop control. Inverters can have different power set-points during grid-connected mode but in stand-alone mode they all need their power set-points to be adjusted according to their power ratings. However, during sudden unintentional islanding (due to loss of mains), transient power can flow from inverters with high power set-points to inverters with low power set-points, which can raise the DC link voltage of the inverters causing them to shut down. This thesis investigates the transient circulating power between paralleled inverters during unintentional islanding and proposes a controller to limit it. The controller monitors the DC link voltage and adjusts the power set-point in proportion to the rise in the voltage. A small signal model of an island microgrid has been developed and used to design the controller. The model and the controller design have been validated by simulation and practical experimentation. The results confirmed the performance of the proposed controller for limiting the DC link voltage and supporting a seamless mode transfer. The limitation of the droop controller, that is utilized to achieve load sharing between parallel-operated inverters in island mode, has also been addressed. Unequal output impedances among the distribution generation (DG) units lead to the droop control being inaccurate, particularly in terms of reactive power sharing. Many methods reported in the literature adopt low speed communications to achieve efficient sharing. However, the loss of this communication could lead to inaccuracy or even instability. An improved reactive power-sharing controller is proposed in this thesis. It uses the voltage at the point of common coupling (PCC) to estimate the inductance value of the output impedance including the impedance of the interconnecting power cables and to readjust the voltage droop controller gain accordingly. In an island microgrid consisting of parallel-connected inverters, the interaction between an inverter’s output impedance (dominated by the inverter’s filter and voltage controller) and the impedance of the distribution network (dominated by the other paralleled inverters’ output impedances and the interconnecting power cables) might lead to instability. This thesis studies this phenomenon using root locus analysis. A controller based on the second derivative of the output capacitor voltage is proposed to enhance the stability of the system. Matlab simulation results are presented to confirm the validity of the theoretical analysis and the robustness of the proposed controller. A laboratory-scale microgrid consisting of two inverters and local load has been built for the experimental phase of the research work. A controller for a voltage source inverter is designed and implemented. A dSPACE unit has been used to realize the controller and monitor the system in real time with the aid of a host computer. Experimental results of the two voltage source inverters outputs are presented.

Identiferoai:union.ndltd.org:bl.uk/oai:ethos.bl.uk:676394
Date January 2015
CreatorsIssa, Walid R. M.
ContributorsAbusara, Mohamad ; Mallick, Tapas
PublisherUniversity of Exeter
Source SetsEthos UK
Detected LanguageEnglish
TypeElectronic Thesis or Dissertation
Sourcehttp://hdl.handle.net/10871/17435

Page generated in 0.002 seconds