Bridging Artificial Intelligence and Power Systems Education Using a Hands-On Executable Framework
A community survey of researchers and practitioners shows 92% report at least one barrier before running an AI model and 94% want a power-specific hands-on course. The paper presents an open, executable module library that maps core AI concepts onto power-system tasks, including DNN templates for load-curve fitting and a CNN power-flow surrogate for a 5-bus system.
A research paper proposes a hands-on executable framework to lower the entry barrier for applying AI in power systems. It addresses the gap between specialized AI applications and reusable educational material, introducing a progressive difficulty ladder of modules that teach foundational DNNs and domain-coupled CNNs for power-flow tasks.
The framework introduces engineering-grounded AI (EGAI), where AI workflows follow power-system domain rules. It provides executable modules with progressive difficulty, starting from DNN function approximation to a CNN surrogate for power-flow on a 5-bus system, enabling learners to build domain-specific models rather than relying on black-box LLMs.
The high demand (94% want a power-specific course) and reported barriers (92%) indicate a significant skills gap in the power industry for AI adoption. This educational framework could accelerate workforce readiness and interdisciplinary collaboration.
By lowering the barrier to AI in power systems, the framework can reduce training costs, speed up prototyping of AI solutions for grid operators, and foster innovation in energy forecasting, optimization, and control.
Next signals include adoption of the framework in university curricula or industry training programs, expansion to larger power systems, and integration with real-world grid data. Potential for open-source community contributions and validation on more complex tasks.