Integrating Generative AI Into Project-Based Learning for Computer Analysis of Power Systems Analysis

Abstract

As generative AI (genAI) tools become increasingly prevalent in higher education, their potential to enhance student learning and engagement in engineering courses warrants focused investigation. This paper presents the redesign of an advanced power systems analysis course that integrates Python programming, project-based learning (PjBL), and generative AI support tools such as ChatGPT. The course challenges students to develop a functioning power systems simulator from scratch while using commercial software (PowerWorld) for validation. We examine the educational impact of this approach through two research questions focused on learning outcomes and student motivation and persistence. Using milestone completion and course grade data across multiple offerings (2023-2025), we observe higher completion rates on the most advanced milestones in 2025, along with reduced performance variability relative to 2024. Student course evaluations remained consistently strong across both years, and open-ended responses indicate that genAI was perceived as a useful support for overcoming programming barriers. This paper outlines the course design, related literature, methodology, and results, offering insights for educators seeking to incorporate AI-assisted PjBL strategies in engineering education.