Journal
MECHANICS OF MATERIALS
Volume 142, Issue -, Pages -Publisher
ELSEVIER
DOI: 10.1016/j.mechmat.2019.103293
Keywords
Residual properties; Post-fire conditions; Material models; Artificial neural network (ANN); Genetic algorithm (GA); Machine learning (ML)
Categories
Ask authors/readers for more resources
Temperature rise, as in the case of fire, severely damages the properties of construction materials and imposes temperature-induced degradations that alter their microstructure and characteristics. As such, practitioners often struggle when assessing residual state of a fire-damaged structure especially due to the lack of insights into residual (post-fire) properties of construction materials. With the hope of narrowing this knowledge gap, this study presents an approach to derive residual material models for a variety of construction materials such as normal strength concrete (NSC), high strength concrete (HSC), ultra-high performance concrete (UHPC), mild steel (MS), high strength steel (HSS), cold formed steel (CFS), stainless steel (SS), glass fiber reinforced polymer (GFRP) and carbon fiber reinforced polymer (CFRP). This approach leverages a hybrid combination of two machine learning (ML) techniques (artificial neural networks (ANN) and genetic algorithms (GA)) to derive specifically tailored material models capable of tracing the post-fire behavior of construction materials. When implemented, the proposed material models could enable proper and unified assessment of fire-damaged structures.
Authors
I am an author on this paper
Click your name to claim this paper and add it to your profile.
Reviews
Recommended
No Data Available