The European Union’s energy and climate policy objectives for 2030 targets to achieve at least 27% of the generated electricity from renewables. Such large-scale integration of renewable Energy Sources (RES) into the grid, will have drastic effects on the electricity grid structure, system operations and the functioning of the electricity market itself. In order to tackle the intermittency challenges posed by RES, smarter operating processes are essential. This requires accurate simulation tools and an efficient exchange of information between the energy players in Europe (TSOs, DSOs and other private generators). The Common Grid Model Exchange Standard (CGMES) was developed to support data exchanges between these energy players. The CGMES based Common Information Model (CGMES-CIM) files contain all the information about the grid under study including the powerflow values. The current version of CGMES-CIM supports full interoperability with respect to steady state simulations but is quite challenging to use for dynamic simulations. Open-Instance Power System Library (OpenIPSL), which is a Modelica based power system library, seems to be one of the possible solutions that can overcome the interoperability issue with respect to dynamic simulations. This thesis focuses on extending the OpenIPSL with PowerFactory based models and provides a proof of concept to automatically initialize the Modelica grid model using CGMES-CIM. OpenModelica, an open-source Modelica simulation environment is used in this project to carry out dynamic studies on a modified Bonneville Power Administration (BPA) grid model. In order to perform dynamic studies, the Modelica grid model first needs to be populated with powerflow values. OpenModelica does not include powerflow option and therefore, powerflow solution needs to be obtained from another simulation tool and the values need to be subsequently loaded into OpenModelica to perform dynamic simulations; this process is called Initialization. Thus, to overcome the challenges with initialization of Modelica models, this thesis presents a proof of concept that directly utilizes the CGMES-CIM files for initialization of Modelica based grid models. The interfacing process between CIM and Modelica based grid model is
done using Python.
To test the concept, Modelica based dynamic simulations were carried out and compared with reference results (signal records obtained through time domain simulations) from PowerFactory. For this purpose, several PowerFactory based Modelica models were developed and validated, extending the OpenIPSL. These models were subsequently used to create and validate a modified BPA grid model, which serves as a test case. This thesis also proposes a method to directly convert CGMES-CIM files to Modelica files. This conversion helps achieve
complete automation of dynamic simulations in a Modelica environment. Therefore, this thesis contributes to the OpenIPSL by developing PowerFactory based models and, also proposes a new CGMES-CIM to Modelica converter.

This paper discusses and compares two approaches to address technical challenges in performing collaborative studies of power system dynamics. On one side, we consider the model migration approach which is an essential piece of dynamic model exchange. On the other side, we look at the co-simulation approach which is used to couple simulation tools together. The approaches are compared in terms of accuracy and the most probable reasons for discrepancy between dynamic responses are discussed. We complement the discussion and conclusions with the simulation results on a simple test system.@en