Research on the automatic BIM modeling method of substation based on LLM agent

Substations, as crucial components of the power system, play a vital role in the design, construction, operation, and maintenance of smart grids. Traditional substation modeling methods often rely on manual operations, suffering from issues such as long modeling cycles, low accuracy, and poor collaboration. However, with the rapid development of information technology, particularly the continuous advancements in artificial intelligence (AI) and Building Information Modeling (BIM) technologies, new solutions for substation modeling have emerged. This paper primarily investigates an architecture that combines Large Language Model (LLM) with intelligent agents, utilizing large models to construct an intelligent agent system. This system is trained to recognize two-dimensional drawings and employs intelligent agents to invoke Revit tools for three- dimensional BIM modeling of substations. Compared to traditional modeling methods, this approach significantly enhances the efficiency and accuracy of substation modeling while reducing modeling costs, promising to accelerate the development of smart grid construction. Therefore, the automatic BIM modeling method for substations using LLM intelligent agents holds broad application prospects and development potential in modern smart grid construction.