| dc.description.abstract | This thesis explores advanced control implementation in Modelica, focusing on two methods: calling external C code via the Modelica external objects class and utilizing Functional Mock-up Units (FMUs) exported from Modelica for Python simulation. The Model Predictive Control (MPC) chapter covers various models and the optimization processes.
Practical implementations involve evaluating an air heater model using first-principles, transfer functions, and state-space models. The thesis integrates a C noise generator into Modelica and utilizes the NLopt library for optimization. Chapters extend the study to Python, illustrating PI-controller simulation using an FMU from OpenModelica and its integration into an MPC framework. The results are presented through Python source code, statistical measures, and visual comparisons.
Serving as a user manual, the thesis provides detailed implementation descriptions for both methods and ensures transparency and accessibility through GitHub. In conclusion, it not only offers insights into advanced control strategies in Modelica but also serves as a practical guide for implementation. | |
