Graph Neural Networks for Learning Nonlinear Power System Operations

Explore the application of Graph Neural Networks in nonlinear power system operations through this comprehensive webinar. Delve into power flow models, optimization problems, and computational benefits while examining topology adaptivity and emergency response strategies. Learn about centralized optimization techniques and discover how graph filters and reinforcement learning contribute to solving complex power system challenges. Gain valuable insights from presenter Hao Zhu of UT Austin as part of the Data sciEnce on GrAphS (DEGAS) Webinar Series, presented in collaboration with the IEEE Signal Processing Society Data Science Initiative.

Introduction
Welcome
Presentation
Power Flow Model
Applications
Topics
Optimization Problem
Computational Benefits
Nonlinear Power Flow Model
Results
Topology Adaptivity
More Results
Emergency Response
Centralized Optimization
Summary
Questions
Graph Filter
Reinforcement Learning