Predicting stress-strain behavior of normal weight and lightweight aggregate concrete exposed to high temperature using LSTM recurrent neural network

• Published in: Construction & building materials 2023-01, Vol.362, p.129703, Article 129703
• Main Authors: Tanhadoust, A., Yang, T.Y., Dabbaghi, F., Chai, H.K., Mohseni, M., Emadi, S.B., Nasrollahpour, S.
• Format: Article
• Language: English
• Subjects: Artificial intelligence; Deep learning; Fire resistance; Lightweight concrete; Long short-term memory
• ISSN: 0950-0618; 1879-0526
• DOI: 10.1016/j.conbuildmat.2022.129703

•240 lightweight and normal weight aggregate concrete were subjected to ambient and elevated temperatures.•The behavior of the lightweight and normal weight aggregate concrete is systematically reported.•An established neutral network, named Long Short Term Memory (LSTM) was used to predict the behavior of the specimen properties under different temperature.•Detailed parameter study on the performance of the LSTM was conducted.•The results demonstrate that the proposed LSTM model can be used to reliable and can effectively predict the.compressive strength, modulus of elasticity, and failure strain of both lightweight and normal weight aggregate concrete. Lightweight aggregate is commonly used to reduce the self-weight of concrete elements which can efficiently mitigate the environmental impact. This study investigates the mechanical performance of normal weight aggregate concrete (NWAC) and Lightweight aggregate concrete (LWAC) exposed to high temperature, focusing on developing a predictive model for the stress-strain relationship of the LWAC based on the property of cement, silica fume, light weight aggregate and water/cement ratio. Mechanical properties of 30 different mixtures, including uniaxial compressive stress–strain relationship, compressive strength, modulus of elasticity, and failure strain are determined before and after high temperature exposures at 250, 500, and 750 °C. Results show that the modulus of elasticity and compressive strength of LWAC increased with exposure temperature. In particular, mixture S4 was found to outperform other mixtures, which retained 96, 75, and 46 % of compressive strength following exposures to 250, 500, and 750 °C, respectively. At all the temperatures investigated, the specimens of this mixture were able to achieve satisfactory modulus of elasticity and maximum strain. The specimens prepared with mixtures containing 75 % LECA (S23) showed a higher retention compressive strength at 750 °C. This study also utilizes the long short-term memory (LSTM) neural network to predict the stress-strain relationship of NWAC and LWAC mixtures after exposure to high temperatures. The results show that the LSTM model could adequately predict the stress-strain relationship of both LWAC and NWAC mixtures in terms of compressive strength, modulus of elasticity, and failure strain at elevated temperatures.