Table 2 Literature Survey for studies on the estimation of concrete materials compressive strength under elevated temperatures using data driven approaches.
From: Compressive strength of nano concrete materials under elevated temperatures using machine learning
References | Models used | Input parameters | Temperatures (°C) | Datasets | Performance | Developing prediction equations |
|---|---|---|---|---|---|---|
Uysal and Tanyildizi (2012)23 | ANNs | Cement, Fly ash, Zeolite, Limestone Powders, Basaltic, Marble Powders, Natural aggregate, group I aggregate, group II aggregate, Polypropylene Fibers, and Temperature | 20–800 | 85 | R² = 0.9757 (all datasets) | ---- |
Tanyildizi (2018)24 | ANNs SVM | Cement, Silica fume, Carbon fiber, Aggregate, and Temperature | 20–800 | 144 | R² = 0.99 (all datasets) | ---- |
Ashteyat and Ismeik (2018)25 | ANNs | Cement, water, fine aggregate, coarse aggregate, water-to-cement ratio, super plasticizer dosage, silica fume, slag, fly ash, marble, basalt, limestone, glass, quartz, polypropylene, steel fiber, relative humidity, and Temperature | 20–900 | 332 | R² = 0.943 (all datasets) | ---- |
Abbas, et al. (2019)26 | ANN | Aggregate/binder ratio, water/binder ratio, soaking period, heating rate, elevated temperature. | 50-1200 | 460 | R² = 0.94 (best predictive model) | ---- |
Ahmed et al., (2021)27 | AdaBoost random forest decision tree | Temp degree; Amounts of cement, water, aggregates, silica fume, nano silica, fly ash, and super plasticizer | 20–600 | 207 | For the three models R² = 0.938, 0.935, 0.911 | ---- |
Alaskar et al. (2023) 28 | GEP and MEP | Nano silica, cement, fly ash, water, silica fume, superplasticizer, sand, gravels, and Temperature | 20-1000 | 207 | R² = 0.854 for training and 0.842 for testing | √ |
Alyousef et al. (2023)29 | ANFIS ANN GEP | Cement content, water, water/binder ratio, coarse aggregate, fine aggregate. | 20-1200 | 307 | R² = 0.937, 0.925,0.996 for the testing models, in order. | ---- |
Kharrazi et al. (2024)30 | convolutional neural network | Mixture proportion, chemical compositions, specimen shape and size, heating regime, and age. | 50-1200 | 1062 | R² = 0.965 (all datasets) | ---- |
Farhangi et al. (2024)31 | ANN | Mix design of concrete, geometrical and mechanical properties of fiber, the maximum heat. | 100–1200 | 286 | R² = 0.988 (best predictive model) | ---- |
