Predicting the Mechanical Properties of RCA-Based Concrete Using Supervised Machine Learning Algorithms

Materials, Journal Year: 2022, Volume and Issue: 15(2), P. 647 - 647

Published: Jan. 15, 2022

Environment-friendly concrete is gaining popularity these days because it consumes less energy and causes damage to the environment. Rapid increases in population demand for construction throughout world lead a significant deterioration or reduction natural resources. Meanwhile, waste continues grow at high rate as older buildings are destroyed demolished. As result, use of recycled materials may contribute improving quality life preventing environmental damage. Additionally, application coarse aggregate (RCA) essential minimizing issues. The compressive strength (CS) splitting tensile (STS) containing RCA predicted this article using decision tree (DT) AdaBoost machine learning (ML) techniques. A total 344 data points with nine input variables (water, cement, fine aggregate, RCA, superplasticizers, water absorption maximum size density RCA) were used run models. was validated k-fold cross-validation coefficient correlation (R2), mean square error (MSE), absolute (MAE), root values (RMSE). However, model's performance assessed statistical checks. sensitivity analysis determine impact each variable on forecasting mechanical properties.

Language: Английский

---

Compressive Strength Evaluation of Ultra-High-Strength Concrete by Machine Learning

Materials, Journal Year: 2022, Volume and Issue: 15(10), P. 3523 - 3523

Published: May 13, 2022

In civil engineering, ultra-high-strength concrete (UHSC) is a useful and efficient building material. To save money time in the construction sector, soft computing approaches have been used to estimate properties. As result, current work sophisticated techniques compressive strength of UHSC. this study, XGBoost, AdaBoost, Bagging were employed techniques. The variables taken into account included cement content, fly ash, silica fume silicate sand water superplasticizer steel fiber, fiber aspect ratio, curing time. algorithm performance was evaluated using statistical metrics, such as mean absolute error (MAE), root square (RMSE), coefficient determination (R2). model's then statistically. XGBoost technique, with higher R2 (0.90) low errors, more accurate than other algorithms, which had lower R2. UHSC can be predicted technique. SHapley Additive exPlanations (SHAP) analysis showed that highest positive influence on strength. Thus, scholars will able quickly effectively determine study's findings.

Language: Английский

---

Machine learning interpretable-prediction models to evaluate the slump and strength of fly ash-based geopolymer

Journal of Materials Research and Technology, Journal Year: 2023, Volume and Issue: 24, P. 100 - 124

Published: March 4, 2023

This study used three artificial intelligence-based algorithms – adaptive neuro-fuzzy inference system (ANFIS), neural networks (ANNs), and gene expression programming (GEP) to develop empirical models for predicting the compressive strength (CS) slump values of fly ash-based geopolymer concrete. A database 245 CS 108 were established from published literature, where 17 significant parameters chosen as input variables development models. The trained tested using statistical measures including Nash-Sutcliffe efficiency, root-squared error, root-mean-square relative-root-mean-square mean absolute correlation coefficient, regression coefficient. comparison results showed that GEP model was superior ANFIS ANN in terms R-value, R2, RMSE both prediction. R-value 0.94 (GEP), 0.92 0.91 (ANN), while it 0.96 0.90 (ANN). Moreover, performance index factor found 0.03 0.029 GEP-models 0.036, 0.030 ANFIS-models 0.035 0.034 ANN-models respectively. sensitivity parametric analysis also performed GEP-developed model. Results demonstrate generates more accurate prediction after being rigorously its hyperparameters optimized.

Language: Английский

---

A critical review on modeling and prediction on properties of fresh and hardened geopolymer composites

Journal of Building Engineering, Journal Year: 2024, Volume and Issue: 88, P. 109184 - 109184

Published: April 3, 2024

Language: Английский

---

A comprehensive overview of geopolymer composites: A bibliometric analysis and literature review

Case Studies in Construction Materials, Journal Year: 2021, Volume and Issue: 16, P. e00830 - e00830

Published: Nov. 30, 2021

Cement is the main component of concrete, a widely used building material. production requires substantial energy, exhausts natural resources, and causes CO2 emissions. Efforts are being undertaken to develop concrete binder instead cement. Geopolymer composite (GPC) developed as potential new material, offering clean alternative for construction sustainability. This study accumulated extensive bibliometric data on GPC from Scopus database conducted scientometric analysis employing an appropriate software. The leading sources publications, highly keywords in published articles, writers papers with highest citations, active regions were all identified analysis. Manual reviews unable deal vast effectively. Additionally, this covers most crucial research concerns shortcomings existing about adoption application. Lastly, further guidelines suggested. review will assist academics various states exchange innovative ideas expertize, encourage collaborative research, joint ventures.

Language: Английский

---

Evaluation of geopolymer concrete at high temperatures: An experimental study using machine learning

Journal of Cleaner Production, Journal Year: 2022, Volume and Issue: 372, P. 133608 - 133608

Published: Aug. 14, 2022

Language: Английский

---

Geopolymer: A Systematic Review of Methodologies

Materials, Journal Year: 2022, Volume and Issue: 15(19), P. 6852 - 6852

Published: Oct. 2, 2022

The geopolymer concept has gained wide international attention during the last two decades and is now seen as a potential alternative to ordinary Portland cement; however, before full implementation in national standards, requires clarity on commonly used definitions mix design methodologies. lack of common definition methodology led inconsistency confusion across disciplines. This review aims clarify most existing diverse procedures methodologies attain good understanding both unary binary systems. puts into perspective crucial facets facilitate sustainable development adoption standards. A systematic protocol was developed based Preferred Reporting Items for Systematic Reviews Meta-Analyses (PRISMA) checklist applied Scopus database retrieve articles. Geopolymer product polycondensation reaction that yields three-dimensional tecto-aluminosilicate matrix. Compared systems, systems contain complex hydrated gel structures polymerized networks influence workability, strength, durability. optimum utilization high calcium industrial by-products such ground granulated blast furnace slag, Class-C fly ash, phosphogypsum or give C-S-H C-A-S-H gels with dense enhance strength gains setting times. As there no standard, designs apply trial-and-error approach, few Taguchi particle packing fraction method, response surface methodology. adopted require optimization certain mixture variables whilst keeping constant other nominal material factors. production NaOH gives less CO2 emission compared Na2SiO3, which higher calcination temperatures Na2CO3 SiO2. However, their usage considered unsustainable due caustic nature, energy demand, cost. Besides blending ash by-products, also use an ingredient blended parameters identified this can help foster robust “go-to” cement construction. Furthermore, proposed future research areas will address various innovation gaps observed current literature view environment society.

Language: Английский

---

Evaluation of Artificial Intelligence Methods to Estimate the Compressive Strength of Geopolymers

Gels, Journal Year: 2022, Volume and Issue: 8(5), P. 271 - 271

Published: April 26, 2022

The depletion of natural resources and greenhouse gas emissions related to the manufacture use ordinary Portland cement (OPC) pose serious concerns environment human life. present research focuses on using alternative binders replace OPC. Geopolymer might be best option because it requires waste materials enriched in aluminosilicate for its production. geopolymer concrete (GPC) is growing rapidly. However, substantial effort expenses are required cast specimens, cures, tests. Applying novel techniques said purpose key requirement rapid cost-effective research. In this research, supervised machine learning (SML) techniques, including two individual (decision tree (DT) gene expression programming (GEP)) ensembled (bagging regressor (BR) random forest (RF)) algorithms were employed estimate compressive strength (CS) GPC. validity comparison all models made coefficient determination (R2), k-fold, statistical assessments. It was noticed that SML performed better than forecasting CS model results also reasonable range. R2 value BR, RF, GEP, DT 0.96, 0.95, 0.93, 0.88, respectively. models' lower error values such as mean absolute (MAE) root square errors (RMSE) verified higher precision ensemble methods. RF (MAE = 2.585 MPa, RMSE 3.702 MPa) BR 2.044 3.180) 4.136 6.256 GEP 3.102 4.049 MPa). application will benefit construction sector with fast methods estimating properties materials.

Language: Английский

---

Compressive Strength of Steel Fiber-Reinforced Concrete Employing Supervised Machine Learning Techniques

Materials, Journal Year: 2022, Volume and Issue: 15(12), P. 4209 - 4209

Published: June 14, 2022

Recently, research has centered on developing new approaches, such as supervised machine learning techniques, that can compute the mechanical characteristics of materials without investing much effort, time, or money in experimentation. To predict 28-day compressive strength steel fiber-reinforced concrete (SFRC), i.e., individual and ensemble models, were considered. For this study, two approaches (SVR AdaBoost SVR bagging) one technique (support vector regression (SVR)) used. Coefficient determination (R2), statistical assessment, k-fold cross validation carried out to scrutinize efficiency each approach In addition, a sensitivity was used assess influence parameters prediction results. It discovered all performed better terms forecasting outcomes. The method most precise, with R2 = 0.96, opposed bagging support regression, which had values 0.87 0.81, respectively. Furthermore, based lowered error (MAE 4.4 MPa, RMSE 8 MPa), tests verified optimum performance AdaBoost. forecast other hand, equally satisfactory. order construction materials, these be applied.

Language: Английский

---

Artificial intelligence-based estimation of ultra-high-strength concrete's flexural property

Case Studies in Construction Materials, Journal Year: 2022, Volume and Issue: 17, P. e01243 - e01243

Published: June 15, 2022

Advancement in Artificial Intelligence (AI) techniques and their applications the construction industry, particularly for predicting mechanical properties of concrete, leads to conservation efforts, time cost. However, insufficient research has been done on Ultra-high-strength concrete (UHSC). For this reason, study aims predict UHSC flexural strength by applying sophisticated AI approaches. Ensembled machine learning performed well compared individual decision tree (DT) model. In current research, is predicted employing supervised Machine Learning (ML) approaches, i.e., DT-Bagging, DT-Gradient Boosting, DT-AdaBoost, DT- XG Boost. Moreover, model performance assessed with help R2, Mean Absolute Error (MAE), Root Square (RMSE). addition that, validation also using k-fold cross-validation technique. Higher R2 0.95 lesser error (i.e., RMSE MAE) values, case DT-Bagging model, depict improved respect other applied ensemble methods. The assessment output shows that anticipated outcomes from proposed models, are much closer actual results experiments, which indicates enhanced prediction UHSC. Further, SHapley Additive exPlanations (SHAP) analysis steel fiber content highest positive influence strength.

Language: Английский

---