P

Power Grid

This thesis addresses the challenges of modernizing Germany's power grid, quantifying risks posed by renewable energy in terms of grid stability. Stability measures are defined. Variability is quantified by fitting Gaussian Mixture Models (GMM) Principal on the Component Analysis (PCA) which is used for dimensionality reduction. Using MCMC techniques find low-probability patterns linked to blackouts and desynchronization.

The code in this repository is used for the research on multi-agent reinforcement learning for power grids of Erica van der Sar. The results in the paper "Multi-Agent Reinforcement Learning for Power Grid Topology Optimization" (https://doi.org/10.48550/arXiv.2310.02605) are produced using this code base.

Binary searched samples of randomly chosen power vectors of a linear grid with lengths 20 to 100.