Search

• A deep learning model was proposed to estimate the temperature exposure of mortar and concrete samples based on their SEM images. • A large dataset of 16,484 SEM images with temperature labels ranging from 20 to 800 °C was collected. • The generalizability of the model was evaluated by testing it on different mix proportions of mortar and concrete. • Visualization was applied to analyze the features learned by the model and to identify the regions of interest for different temperature levels. This study presents a framework for identifying the temperature experienced by fire-damaged mortar and concrete using scanning electron microscopy (SEM) images and deep learning. A dataset of 16,484 SEM images of cement paste mixes with varying water-to-binder ratios and pozzolanic materials, exposed to temperatures ranging from 200 to 800 °C was established. Then a deep-learning model based on a convolutional neural network (CNN) for SEM image classification was trained, achieving a high accuracy above 98 %. To test the method's generalizability, cement paste mixture with a different water-to-binder ratio, mortar mixture with sand inclusion, and concrete mixture with coarse aggregates were prepared and exposed to different temperatures. The predicted temperatures deviated from the target temperatures within 8.6 %. Finally, visualization of the deep learning model was used to identify the critical features that influenced the prediction. The outer hydration products with smaller pores had a higher influence on samples before heating, whereas porous dehydrated products were more influential in samples exposed to high temperatures. [ABSTRACT FROM AUTHOR]

• Compressive strength of both UHPC and mortar increases at 300 ℃ but decreases at higher temperature. • Post-fire water curing is more effective on recover the compressive strength of the UHPC mix and Mortar than the post-fire air curing. • Strength recovery of the Mortar is greater than that of the UHPC. • Precipitation of portlandite heals the coarsened microstructure and recovers compressive strength. This paper investigates effects of post-fire curing on strength recovery and changes in microstructure of ultra-high performance concrete (UHPC) and mortar. Specimens were heated to elevated temperatures 300, 600, and 900 °C with a heating rate of 1 °C/min. After cooling, the specimens were subjected to post-fire curing in either air or in water for 56 days. Compressive strength, X-ray diffraction (XRD), mercury intrusion porosimeter (MIP), and field emission scanning electron microscope (FESEM) tests were conducted to examine the changes at both macro and micro scale and to reveal deterioration mechanisms due to fire effect and healing mechanisms due to post-fire curing. The test results showed that compressive strength of both UHPC and mortar increased slightly after 300 °C exposure but decreased significantly at higher temperature due to the coarsening of microstructure. Post-fire curing in air showed very limited strength recovery since new curing products did not heal thermal cracks larger than 1 μm. On the other hand, curing in water led to substantial strength recovery of the mortar mix due to formation of portlandite which compactly filled the microcracks in fire-damaged samples. However, the loose ettringite phase formed in the UHPC mix did not contribute to regaining the initial compressive strength. [ABSTRACT FROM AUTHOR]