Overview and details of the sessions of this conference. Please select a date or location to show only sessions at that day or location. Please select a single session for detailed view (with abstracts and downloads if available).
This session delves into the transformative potential of generative AI (GenAI) for building energy modeling (BEM) and HVAC system design. Presentations will explore how GenAI tools are revolutionizing professional learning and design workflows by automating tasks such as EnergyPlus model development, system sizing, fault diagnostics, and data management; and highlight emerging applications for AI-assisted tutoring systems, and scenario-based learning for engineers. The session also introduces an open-source platform to generate complex BEMs through natural language prompts, significantly reducing manual modeling efforts. Attendees will gain a forward-looking perspective of how GenAI can streamline energy modeling processes, elevate professional training, and make sophisticated modeling tools more accessible to practitioners.
Presentations
Transforming Education and Workforce Training in Building Energy and HVAC with GenAI
Liang Zhang
University of Arizona
This presentation explores how GenAI technologies can automate and enhance professional learning across multiple dimensions, including building energy modeling workflows (e.g., EnergyPlus model development, debugging, and calibration), HVAC system design and operation (e.g., load calculation, system sizing, and fault diagnostics), and broader data management skills (e.g., data cleaning, transformation, and knowledge retrieval). Specific training applications include intelligent tutoring systems for energy modelers, AI-assisted certification exam preparation, hands-on scenario generation for HVAC engineers, and personalized feedback for building energy analysts.
An Open-Source Automated Platform for Complex Building Energy Modeling from Natural Language
Gang Jiang
The University of Utah
Building energy modeling (BEM) often requires significant manual effort, limiting its widespread application in design and operational decision-making. To address this challenge, this study proposes EPlus-LLMv2, an open-source, large language model (LLM)-based platform that enables users to automatically generate complex building energy models using natural language. The proposed method enables highly customized LLM-based BEM for complex cases while maintaining computational efficiency. An interactive human-AI interface is also implemented to improve usability for practitioners.
