A hierarchy of microgrid models with some applications

Abstract

This dissertation proposes a hierarchy of microgrid models that can be utilized for power system analysis and control design purposes. The microgrid models are classified according to time resolution and, consequently, according to complexity and computational cost as well. Our approach involves three key stages: (1) the formulation of high-order models, using existing circuit and control laws in the literature, followed by (2) a systematic reduction of the high-order models to reduced-order models using singular perturbation analysis and, finally (3) a hierarchical classification of the high-order and reduced-order models according to the time-scales we propose they should be utilized for. The resulting hierarchy of microgrid models is comprised of a microgrid high-order model (µHOm), the microgrid reduced-order model 1 (µROm1), the microgrid reduced-order model 2 (µROm2), and the microgrid reduced-order model 3 (µROm3), the last three of which are developed in this work.

Each microgrid model we develop is composed of models for three-phase inverters, microturbines, type-C wind turbine generators, distribution line networks, and generic elements (e.g. loads) connected to the network. We identify the time resolution of the models and analyze the performance of these models using various power system test cases.

We also showcase two applications of our reduced-order models: first in the design of a robust synchronization method for electric power generators, and secondly in the development of a hardware-in-the-loop laboratory for analysis and testing of microgrid controls.