Master Thesis Student
Tlk-Thermo Gmbh And Institut Für Thermodynamik [Institute Of Thermodynamics] At Tu Braunschweig
Braunschweig Area, Germany
🔵 Thesis: Robust parameter-optimization of dynamic thermal systems with Functional Mock-Up Interface🔸 Implemented two different derivative-free optimization algorithms in Python- Implemented globalized constrained bounded Nelder Mead method- Implemented a self-designed genetic algorithm🔸 Used method from data science to solve a specific problem in numerical optimization in Python- Implemented the algorithm of principal components analysis to verify the degenerated level… Show more 🔵 Thesis: Robust parameter-optimization of dynamic thermal systems with Functional Mock-Up Interface🔸 Implemented two different derivative-free optimization algorithms in Python- Implemented globalized constrained bounded Nelder Mead method- Implemented a self-designed genetic algorithm🔸 Used method from data science to solve a specific problem in numerical optimization in Python- Implemented the algorithm of principal components analysis to verify the degenerated level of a simplex- Used probabilistic restarts to search for global minimums🔸 Studied Lagrange Multipliers method and different penalty functions (for instance, augmented Lagrangian algorithm) to handle the constraints 🔸 Validated the self-implemented optimization algorithms using FMU interface and Modelica programming- Studied the principle (solving process of DAEs) of FMU- Modeled a discretized transient thermal system to simulate the heat transfer in a braking disc and optimized the geometry parameter of the brake disc🔸 Studied the advantages and disadvantages of the above-mentioned two derivative-free algorithms (efficiency, sensibility to steep functions, performances for seeking local and global minimums) Show less