At the Institute of Applied Informatics, methods for modeling, simulation and analysis of electrical grids are developed amongst others. One of the main topics is the development of scalable algorithms for the simulation of power system dynamics in complex and large integrated networks. To increase simulation efficiency, the high performance computing platform ForHLR II at KIT-SCC is to be used together with Julia, which is a high-level, high-performance dynamic programming language for technical computing.
Within the scope of a student thesis, a parallel software for power grid dynamics simulation is to be developed. The main tasks in this work will include:
Studying and analyzing the available Matlab-based sequential computation algorithms for dynamic simulation.
Porting of the Matlab code to the programming language Julia.
Analysis of suitable numerical libraries for parallel DAE solvers.
Development of parallel algorithms using the HPC infrastructure ForHLR II.