Advanced Modeling, Design, and Control of ac-dc Microgrids

An interconnected dc grid that comprises resistive and constant-power loads (CPLs) that is fed by Photovoltaic (PV) units is studied first. All the sources and CPLs are connected to the grid via dc-dc buck converters. Nonlinear behavior of PV units in addition to the effect of the negative-resistance CPLs can destabilize the dc grid. A decentralized nonlinear model and control are proposed where an adaptive output-feedback controller is employed to stabilize the dc grid with assured stability through Lyapunov stability method while each converter employs only local measurements. Adaptive Neural Networks (NNs) are utilized to overcome the unknown dynamics of the dc-dc converters at Distributed Energy Resources (DERs) and CPLs and those of the interconnected network imposed on the converters. Additionally, the use of the output feedback control makes possible the utilization of other measured signals, in case of loss of main signal, at the converter location and creates measurement redundancy that improves reliability of the dc network. The switching between measurement signals of different types are performed through using the NNs without the need to further tuning. Then, in a small-scale ac grid, PV-based Distributed Generation (DG) units, including dc/dc converters and inverters, are controlled such that mimic a synchronous generator behavior. While other control schemes such as Synchronverters are used to control the inverter frequency and power at a fixed dc link voltage, the proposed approach considers both the dc-link voltage and the inverter ac voltage and frequency regulation. The dc-link capacitor stores kinetic energy similar to the rotor of a synchronous generator, providing inertia and contributes to the system stability. Additionally, a reduced Unified Power Flow Controller (UPFC) structure is proposed to enhance transient stability of small-scale micro grids. The reduced UPFC model exploits dc link of the DG unit to generate appropriate series voltage and inject it to the power line to enhance transient stability. It employs optimal control to ensure that the stability of the system is realized through minimum cost for the system. A neural network is used to approximate the cost function based on the weighted residual method