SEMANTIC INFORMATION AND PHYSICAL MULTI-DOMAIN MODELING AND SIMULATION FOR POWER SYSTEMS

Abstract: There are different reasons for combining different modeling languages and simulationlanguages: Exchange of more detailed information about power network components, theirparameters and, most importantly, mathematical equations describing their behavior andthe exchange of a mathematical description, using equation-based languages (e.g.:Modelica), allows models to be detached from the mathematical solver. This leads to thedevelopment of new APIs within software tools, which can handle standardized modelinglanguage used for model implementation. Furthermore, the mathematical description ofmodels and the integration of new simulation standards, such as the FMI, could helpavoiding ambiguities on how power system models are implemented, by providingadditional means for the exchange of the complete description of models or parts of amodel between software tools.The aim of this thesis is to provide a new approach for the development of powersystems modeling and simulation software tools. The thesis is focused on proposing newmethods, based on available information and simulation standards for the exchange,modeling, and simulation of power systems dynamic models; and to show a proof ofconcept of the feasibility of the proposed methods. To this aims, the Common InformationModel (CIM) for the modeling and exchange of power system information is studied.Furthermore, the equation-based language Modelica is described and proposed with theaim of complementing the use of the CIM for the modeling and simulation of thosedynamics models.The application of these standards lead to a different view on the modeling andsimulation of power dynamic network models. The conventional view is that of black boxmodeling. The implementation of network model components is strongly connected to thesimulation software tool used for steady-state and dynamics calculations. Thus, a modeleror test engineer only has access to the parameters of a model and relies on the softwarecapabilities to calculate the states and the behavior of that model. This thesis proposes adifferent view per the application of the white box modeling and simulation concept: fulldetail and transparency on the development of a mathematical description of power systemcomponents and discrete events. Moreover, the combination of information standards withequation-based standards to produce network models allows full access and manipulationof the complete model details. Finally, transparency regarding the implementation ofsoftware tools can support either information-based, equations-based languages orsimulation standards, which are suitable for simulation of dynamic network models.