Bridging Artificial Intelligence and Power Systems Education Using a Hands-On Executable Framework

Synthesis: Junjie Yin, Buxin She, Xinyu Feng, Fangxing Li — arXiv (cs.AI / eess.SY) preprint, 2026 (University of Tennessee, IEEE PES).

Synthesis

Community survey of researchers and practitioners: 92% report at least one barrier before running an AI model and 94% want a power-specific hands-on course.

Framework is a progressive difficulty ladder mapping core AI concepts onto representative power-system tasks: DNN function approximation/load-curve fitting, domain-coupled CNN power-flow surrogate (5-bus), DNN-assisted optimization, DRL for battery storage control, and PINNs for the swing equation.

All modules are Jupyter notebooks running locally or on Google Colab, delivered through an IEEE online course and IEEE PES webinar series.

Webinar drew 590+ live attendees (top-10 most-attended IEEE PES webinars) and 344+ repository visits within two weeks.

Argues for engineering-grounded AI (EGAI): AI workflows should follow established engineering and power-system domain rules rather than acting as task-agnostic black boxes, supporting professional training and curriculum design.

What this means for practice

• Instructors. Structure AI instruction as a difficulty ladder — function approximation, surrogate modeling, optimization, control, physics-informed learning — so each method arrives attached to a power-system task rather than a generic dataset.
• Instructors. Have students run and modify the notebooks locally or in Colab instead of reading them, addressing the barrier the community survey found most binding: 92 percent of the 52 respondents reported at least one obstacle, chiefly getting a model to start (48.1 percent) and hardware limits.
• Designers. Pair every model with the domain rule it serves — a 5-bus power-flow surrogate, DRL for battery storage control, a PINN for the swing equation — so the engineering constraint in curriculum design stays visible and verifiable.
• Faculty developers. Reuse the webinar-plus-notebook format for professional training; 590 or more live attendees and 344 or more repository visits within two weeks indicate the format reaches and engages practitioners.

Limitations

• The modules are stated to be intentionally lightweight, with small demonstration settings chosen for speed rather than realism, so they teach workflow patterns rather than scale to realistic grid systems.
• No formal classroom evaluation of learning outcomes was conducted; the evidence is a targeted community survey, webinar attendance, and repository-visit counts, and the authors name a controlled classroom study as future work.
• The motivation survey rests on 52 self-selected respondents from the IEEE power and energy community, so the reported barrier rates describe that professional population rather than educators broadly.

Citation

Junjie Yin, Buxin She, Xinyu Feng, Fangxing Li (2026). Bridging Artificial Intelligence and Power Systems Education Using a Hands-On Executable Framework. arXiv (cs.AI / eess.SY) preprint.