Equation-based object-oriented modeling, simulation, analysis and control of electric power systems

Thumbnail Image
de Castro Fernandes, Marcelo
Issue Date
Electronic thesis
Electrical engineering
Research Projects
Organizational Units
Journal Issue
Alternative Title
Electric power systems are complex by nature and, therefore, engineers have developed var- ious techniques to understand, characterize and study electrical systems and their dynamic behavior. Throughout history, modeling and simulation tools have been created to provide means to perform meaningful and reliable studies that allowed us to move further in the development of the electric grid. As these systems became more interconnected, their representations increased in complexity, forcing legacy tools to be adapted and enhanced, and new techniques to be developed in order to meet the demand for the study of new use cases. The fact that the current power system is undergoing a major transition, in different aspects, implies the existence of an opportunity for the development and assessment of new tools that might allow us to study future electric power systems. In this context, the present work aims at assessing an equation-based, object-oriented modeling language, namely Modelica, as an alternative to appropriately and accurately representing the complexity of future power system models. This work assesses and investigates the usage of Modelica in the study of power systems in different aspects. First, it describes phasor-based power system models in Modelica, and how they can benefit from equation-based and object-oriented paradigms. Modelica models are used from off-line simulations, to controller design procedures, to real-time simulations, with minimal manipulation, showing a great model reusability, tractability and portability. These results bring to light how Modelica representations of power systems can result in a transparent framework that enables different benefits in modeling, simulation and analysis of these complex systems. In addition, model transformation tools are developed in this work, establishing a bridge between Modelica and traditional simulation tools and model representations that are commonly used in the industry. The model transformation tools, which are shown to be scalable, allow an even greater portability and reusability of power system models. Second, this work also assesses the ability of Modelica in the modeling, simulation and analysis of electromagnetic transients. This is especially relevant in the present scenario, in which power electronics have become more present in ground-based power systems or in smaller networks such as microgrids and electrified vehicles, such as more electric aircrafts. Component models and test systems, representing different turboelectric aircraft propulsion architectures, are developed to assess Modelica in this context. The object-oriented equation-based language is shown to present benefits, especially in the dynamic assessment of systems that might benefit from multi-domain representations or that require not just simulation, but linear analysis to be performed. Finally, this work also presents interfaces that are used to unify, at the equation level, systems that have been constantly considered to be a separate entity from the bulk power system, such as distribution systems or areas with high penetration of inverter-based resources. These interfaces are tested using different power system models assembled in the Modelica environment, and they provide the means to perform the analysis of heterogeneous power system models, as a whole. The strategy is shown to be promising, especially considering the present energy transition context.
School of Engineering
Full Citation
Rensselaer Polytechnic Institute, Troy, NY
Terms of Use
PubMed ID