I maintain an open GitHub-based resource for Fundamentals of AI for Power Electronics: a structured path of Jupyter notebooks, data, and documentation aimed at practitioners and students who want hands-on skill at the intersection of machine learning and power electronics—not only slides, but runnable code you can adapt.
The repo accompanies the invited tutorial article Fundamentals of Artificial Intelligences for Power Electronics (IEEE Transactions on Industrial Electronics, 2026). Module folders are mapped to paper sections so you can read a topic, then open the matching notebooks. There is also a companion education piece (pilot course at the University of Arkansas) in the repo’s docs/ folder on reforming power-electronics education in the AI era.
What you’ll find: a progressive curriculum—from environment setup and metaheuristic optimization, through classical ML, ensembles, neural networks (including 3D thermal-field regression and sequence models), physics-informed modeling (PINN), reinforcement learning on converter examples, simulation automation (e.g. LTspice / PLECS workflows), to applied case studies on buck, DAB, IGBT health, and magnetic modeling. Most notebooks include Open in Colab flows; one automation-heavy module expects local simulators.
Companion tools: an interactive AI-for-PE algorithm selector to narrow methods to your task, plus a documented ChatGPT-based tutor path in the repository README for deeper Q&A aligned with the material. Content is actively refined; code is Apache-2.0 and educational text is CC BY-NC 4.0 per the repo license files.
Fundamentals_of_AI_for_PE on GitHub