Machine learning tool could speed up fire safety assessments for steel beams
Civil EngineeringMaterials ScienceEngineering
THE AI ANGLE
Predicting fire-induced temperatures in protected steel beamsResearchers have developed an automated machine learning framework that rapidly predicts the fire performance and heat transfer of three-sided protected steel beams, completing most assessments in under 60 seconds. Trained on hundreds of automated simulations spanning various beam depths and insulation configurations, the gradient-boosting model achieved a root mean square error of just 1.34°C compared to test data. For engineering faculty, this advance provides an efficient alternative to computationally expensive numerical simulations and parameter-limited analytical equations in structural fire safety design.
THE TEACHING ANGLE
Instructors can explore the trade-offs between computationally intensive finite element modeling and rapid machine learning surrogates when evaluating safety-critical thermal and structural performance.Read the original at techxplore.com Generate teaching or study materials
More in Civil Engineering
- Researchers develop two-level AI framework for more informative steel bridge corrosion inspectionPhys.org — Technology · September 16, 2026
- The biggest issues with delivery robots are exactly what you'd thinkEngadget · September 15, 2026
- AI-powered inspection system gives 3D printers 'a brain behind the eyes'Phys.org — Technology · September 15, 2026