In the quest for sustainable construction materials, researchers have turned to recycled aggregate concrete (RAC) as a promising alternative to traditional concrete. However, the behavior of RAC under elevated temperatures has remained a complex puzzle, until now. A groundbreaking study led by Morteza Ghodratnama from the Department of Civil Engineering at Ferdowsi University of Mashhad in Iran has introduced a novel modeling framework that could revolutionize the way we understand and utilize RAC in high-temperature environments.
The study, published in ‘Case Studies in Construction Materials’ (a journal that translates to ‘Case Studies in Building Materials’ in English), employs Kolmogorov–Arnold Networks (KANs) to predict the thermo-mechanical properties of RAC under elevated temperatures ranging from 150 to 1000°C. This approach marks a significant departure from conventional machine learning methods, as it intelligently segments the temperature spectrum to capture the distinct degradation mechanisms at different thermal regimes.
“Unlike traditional methods that treat the entire temperature range uniformly, our approach acknowledges the unique behaviors of RAC at different temperature intervals,” Ghodratnama explains. “This allows us to provide more accurate predictions and a deeper understanding of how RAC performs under extreme conditions.”
The research utilized a curated database of 260 experimental results from 15 international studies, systematically classified into specific temperature intervals for compressive strength and elastic modulus. The models employed five key predictors: water-to-powder ratio, pozzolanic materials-to-aggregate ratio, recycled aggregate percentage, ambient compressive strength, and the ratio of test to ambient temperature.
The results were impressive, with the optimal KAN models achieving determination coefficients (R²) of up to 0.971. The study found that residual compressive strength declined significantly from approximately 29 MPa in the 150–300°C range to 7.2 MPa in the 600–1000°C range, while the elastic modulus decreased by about 63% beyond 500°C.
Parametric analysis confirmed that temperature ratio and recycled aggregate content were the dominant factors influencing thermal degradation. This finding has profound implications for the energy sector, where the demand for sustainable and fire-resistant construction materials is on the rise.
“Our KAN-based approach offers a high-fidelity, interpretable, and computationally efficient tool for predicting the thermo-mechanical behavior of RAC under elevated temperatures,” Ghodratnama states. “This can provide valuable insights for the design of safer and more sustainable fire-resistant concrete structures, which are crucial for the energy sector.”
The study’s innovative use of KANs and temperature-based segmentation addresses critical gaps in existing literature and sets a new standard for modeling the thermal degradation of RAC. As the construction industry continues to embrace sustainable practices, this research paves the way for the wider adoption of RAC in high-temperature applications, ultimately contributing to a more environmentally friendly and resilient built environment.
In the words of Ghodratnama, “This is not just about understanding the behavior of RAC; it’s about shaping the future of sustainable construction.” With this groundbreaking research, the future of RAC looks brighter and more promising than ever before.