A. Hickey, C. O’Faolain and P. Cuffe, “Large Language Models in Power Engineering Education: A Case Study on Solving Optimal Dispatch Coursework Problems”, presented at IEEE International Conference on IT in Higher Education and Training, Paris, France, November 2024

Powerful chatbots, based on large language models, have recently become available for consumer use. The ability of such chatbots to provide credible textual responses to sophisticated engineering problems has been demonstrated in various subfields. This paper seeks to contribute to this growing body of literature by examining the extent to which such a chatbot can be prompted to complete an archetypal homework exercise for a university-level course in power system operation, specifically the formulation and solution of an optimal dispatch problem. In this modest case study, we employ ChatGPT Plus with the Wolfram plug-in, presenting it with a series of tasks ranging from simple linear programming to complex economic dispatch problems. Our methodology involves providing the chatbot with detailed prompts mirroring typical assignment instructions. Initial results suggest that ChatGPT can successfully solve the linear programming problem, produce a graphical solution, formulate the economic dispatch problem mathematically, derive the Lagrangian function, and generate appropriate GAMS code for solving the optimization problem. While not definitive, these findings indicate that current large language models may be capable of completing some advanced power system engineering coursework, potentially raising important considerations for assessment integrity in engineering education. Further research is needed to more fully understand the implications of these emerging technologies in educational contexts.