AI-assisted thermal and airflow simulation
Learning objectives
By the end of the course, participants will be able to:
- Identify the stages of a thermal and airflow study where AI brings a real gain, and those where it carries a risk
- Prepare and check simulation input data (geometry, occupancy schedules, weather data) with an AI assistant
- Build a simple surrogate model to explore many variants quickly before detailed simulation
- Analyse and summarise simulation results (summer comfort, loads, airflow rates) with the help of AI
- Write a clear calculation note while keeping the assumptions traceable
Audience
Engineers and technicians in thermal, HVAC and fluids consultancies; energy analysts.
Prerequisites
Experience with a dynamic thermal simulation or HVAC sizing tool; basic knowledge of heat transfer. No programming skills required.
Day-by-day programme
| Session | Content |
|---|---|
| Day 1 morning | Overview: generative AI, machine learning and physical simulation; what AI can and cannot do in thermal and airflow engineering |
| Day 1 afternoon | AI-assisted input data: extraction from drawings and specifications, occupancy schedules, weather files; consistency checks |
| Day 2 morning | Dynamic thermal simulation: automating variants (insulation, solar shading, ventilation) and processing results with Python and an AI assistant |
| Day 2 afternoon | Surrogate models: training on a design of experiments, measuring the error, using it to pre-screen variants |
| Day 3 morning | Airflow: flow rates, air change, reading ventilation simulation results; benefits and limits of AI for sizing support |
| Day 3 afternoon | Workshop on a provided case or a participant’s project: variant study, summary of results and calculation note; group review |

