Skip to content

Latest commit

 

History

History
43 lines (24 loc) · 1.7 KB

File metadata and controls

43 lines (24 loc) · 1.7 KB

Workshop AI for Intelligent Energy Systems

The AI for Intelligent Energy Systems Workshop is a 3-day workshop hosted by the Delft AI Energy Lab. The workshop explores the use of AI algorithms to solve emerging challenges in energy systems. Codes from the workshop have been provided for students and researchers. The workshop repository is organized as follows:

Large Language Models (LLMs) in Energy Markets

LLM_for_energy_market.ipynb

  • Interacting with OpenAI API
  • Introduction to Prompt Engineering
  • Using role-playing for LLMs
  • Training your own LLM with Llama and Stable Beluga2
  • Fine-tuning locally trained LLM to predict electricity price

Graph Neural Networks for Power Systems

GNN_for_power_systems

  • Introduction to graphs with pytorch_geometric and networkx
  • Solving common graph tasks with graph neural networks (GNN)
  • Solving Optimal Power Flow (DCOPF) problem with GNN

Deep Reinforcement Learning for Power Systems

DRL_for_power_systems

  • Q-learning for Taxi-driver environment
  • Deep Q networks for cart-pole environment
  • DDPG for mountain car environment
  • Training an intelligent battery storage system

For questions, please contact dai-energy-lab@tudelft.nl

License

This work is licensed under a License: MIT