Simplifying The Configuration Process For Iec 61850-Based Systems Using Generative Ai Technology

Since its inception, IEC 61850 aimed to give protection and control (P&C) engineers involved in the design and application of digital substations the flexibility and clarity in how the different IEDs in the switchyard will interact among each other. This flexibility, however, brought an organizational and executive complexity that hold off many engineers. The perspective of dealing with tens of documents, each one explaining in depth an object-oriented approach to the design and usage of IEDs can challenge many P&C engineers and result in them avoiding utilization of the standard. With the development and popularity of the AI (Artificial Intelligence) in the last few years, it was natural move to use this technology in such processes, serving both as tool to generate and to capture defective configurations. This paper will explore the use of AI, and particularly a Large Language Model (LLM) that understands and generates human-like text by learning patterns from vast amounts of text data. It lays the foundation to help architect projects, where we provide the typical P&C Engineers a system where their questions and requests about IEC 61850 applications are correctly organized and addressed using specifically written prompts and inputs that comes directly from the IEC Standards and device-specific white papers. These input files, despite being lengthy and complex, are easily organized and comprehended by the system guide the Large Language Model (LLM) to deliver to the user a clear answer, analysis or even a whole file compliant to the standard.