Integrated Control of Parallel Connected Multiple SEPIC Converters For DC Microgrid Operation

Abstract

DC microgrids are gaining popularity now a days due to their advantages such as improved efficiency, reliability, and renewable energy integration and also due the environmental challenges like carbon and other greenhouse gases emission from fossil fuel consumption for energy purposes. One of the core challenges in DC microgrids is stabilization and regulation of DC bus voltage, which is crucial for the proper operation of the loads connected to DC bus. Parallel connection of DC DC converters is a commonly used approach for combining and distributing multiple sources of DC power in a microgrid. In this thesis, a fourth order DC DC Single Ended Primary Inductor Converter (SEPIC) converter is selected and it’s mathematical model is built including it’s peracetic resistances/ESRs in order to realize the real case scenarios, based on this mathematical model a Linear Quadratic Controller (LQR) is designed for an optimal control strategy of the parallel connected SEPIC converters in a DC microgrid. The proposed optimal control strategy uses a decentralized distributed control approach, where each converter self regulates its voltage output based on estimated values of all states through a state observer. In this work a decentralized and distributed an optimal control scheme is selected, analyzed and modelled for power balance operations in an islanded 80-100V-LVDC microgrid, consisting of a renewable energy source, an electronic load and storage capability. Control is realized through DC bus voltage monitoring and control operations on the interfacing converters, based on predefined voltage set points. To cope with the converter uncertainties, external disturbance, to annihilate the need for sensors and most importantly to annihilate the bandwidth communication lines a robust optimal controller based on extended state observer (ESO) is proposed and applied to SEPIC converters in parallel mode of operation. The comparison with PI control shows that the proposed method can achieve better disturbance rejection ability without overshoot in step response. The control strategy is designed to provide rapid response to different loading conditions, improve regulation of voltage, and reduce voltage ripples. Through MATLAB/Simulink simulations and hardware module results shows that the proposed control strategy effectively regulates and tracks the DC bus voltage reference and hence ensures the stability of overall DC microgrid under different loading conditions and as well as different reference voltage levels. The proposed control strategy can be a useful approach for DC microgrid designers and operators to ensure the proper functioning of the microgrid and improve its overall performance.