| Title | Carbonation depth prediction and parameter influential analysis of recycled concrete buildings |
|---|---|
| ID_Doc | 9083 |
| Authors | Wang, DC; Tan, QH; Wang, YR; Liu, GY; Lu, Z; Zhu, CQ; Sun, BC |
| Title | Carbonation depth prediction and parameter influential analysis of recycled concrete buildings |
| Year | 2024 |
| Published | |
| Abstract | With the development of the circular economy and low-carbon society, the large-scale application of construction solid waste in buildings, such as recycled concrete, is becoming imperative. Accurately predicting the carbonation depth of recycled concrete is of great significance. Quantitatively analyzing the impact of each parameter on carbonation and elucidating the relationships between these parameters present challenges in predicting the carbonation of recycled concrete. In this study, different machine learning models and prediction equation models were applied and compared to predict the carbonation depth of 576 datasets associated with recycled concrete. The machine learning models used include Automation Machine Learning (AutoML), LightGBM, CatBoost, Neural Networks, Extra Trees, Random Forest, XGBoost, KNN (K-Nearest Neighbor). The results indicate that the machine learning method shows higher accuracy than the traditional equation, the AutoML model exhibits the best prediction accuracy among the investigated machine learning models, and carbonation test results further verified the favorable carbonation depth prediction effects of AutoML model. Furthermore, SHAP (Shapley Additive Explanations) was utilized to quantitatively analyze and explain the prediction results. The results demonstrate that carbonation time and the water to cement (W/C) ratio of recycled concrete have the most significant impact on the carbonation depth of recycled concrete buildings. |
Similar Articles
| ID | Score | Article |
|---|---|---|
17780
|
Dzieciol, J; Sas, W Perspective on the Application of Machine Learning Algorithms for Flow Parameter Estimation in Recycled Concrete Aggregate(2023)Materials, 16, 4 | |
| 66412
|
Al-Fakih, A; Al-wajih, E; Saleh, RAA; Muhit, IB Ensemble machine learning models for predicting the CO2 footprint of GGBFS-based geopolymer concrete(2024) | |
| 9346
|
Maged, A; Elshaboury, N; Akanbi, L Data-driven prediction of construction and demolition waste generation using limited datasets in developing countries: an optimized extreme gradient boosting approach(2024) | |
| 16889
|
Rakhshan, K; Morel, JC; Daneshkhah, A Predicting the technical reusability of load-bearing building components: A probabilistic approach towards developing a Circular Economy framework(2021) | |
| 15649
|
Qin, X; Kaewunruen, S Eco-Friendly Design and Sustainability Assessments of Fibre-Reinforced High-Strength Concrete Structures Automated by Data-Driven Machine Learning Models(2023)Sustainability, 15, 8 |