Smart Grid Assistive AI in Requirement Engineering: Improving the Modeling of Use Cases and Architecture Models with LLMs

The IEC 62559 Use Case Methodology and Smart Grid Architecture Model (SGAM) Framework are crucial for fostering a shared understanding in the development of ICT-based power systems and their components within energy-related Requirements Engineering.However, discrepancies in interpretation among diverse stakeholders often diminish the quality of IEC 62559 Use Cases and SGAM Models, potentially leading to errors and costly setbacks in later project stages.This paper introduces the Smart Grid Assistive AI in Requirements Engineering (SGAAIRE), an intelligent system that leverages Large Language Models to enhance the quality of IEC 62559 Use Case descriptions and SGAM Models.Through a demonstration scenario featuring built-in quality defects, SGAAIRE is shown to effectively address common issues as an extension to a Use Case Management Repository.The proposed system supports the improvement of IEC 62559 Use Cases and SGAM Models, enabling the development of third-party tools aimed at advancing research in AI for energy-related requirements management.