Seminar 8 -- Accelerating Thermal System Simulation with AI

2024 ASHRAE Annual Conference

This product is a zip file that contains files that consist of PowerPoint slides synchronized with the audio-recording of the speaker, PDF files of the slides, and audio only (mp3 format) as noted.

This session explores several ways in which AI and surrogate models enhance model-based design, optimization and control of thermal systems. The main objective is to highlight how surrogate modeling techniques, when applied to HVAC equipment, can greatly reduce the computational costs associated with complex physics-based models. This reduction in computational cost opens up new possibilities for design exploration and analysis that were previously unattainable. Additionally, the session offers valuable insights into the challenges and best practices of applying AI techniques in applications that typically lack extensive data.

1. Using Random-Forest Machine Learning to Model a Direct Expansion Heat Pump for Model Predictive Control Zheng O'Neill, Ph.D., P.E., Texas A&M University, College Station, TX
2. Using Metamodeling Techniques to Accelerate the Evaluation for Thermal Storage Components/Systems Ransisi Huang, PhD, National Renewable Energy Lab; Jason Woods, PhD, National Renewable Energy Laboratory, Washington, DC
3. Harnessing the Power of Machine Learning for Duty Cycle Simulation of Transport Refrigeration Units Rohit Dhumane, PhD, Trane Technologies, Davidson, NC