AutoGluon-enabled machine learning models for predicting recycled aggregate concrete’s compressive strength

The properties of concrete containing coarse recycled aggregates were investigated. Laboratory trials were conducted to investigate the possibility of using recycled aggregates from the demolition wastes available locally as the replacement of natural coarse aggregates in concrete. A series of tests were carried out to determine the density, compressive strength, split tensile strength, flexural strength and modulus of elasticity of concrete with and without recycled aggregates. The water cement ratio was kept constant for all the mixes. The coarse aggregate in concrete was replaced with 0%, 20%, 40%, 60%, 80% and 100% recycled coarse aggregates. The test results indicated that the replacement of natural coarse aggregates by recycled aggregates up to 40% had little effect on the compressive strength, but higher levels of replacement reduced the compressive strength. A replacement level of 100% causes a reduction of 28% in compressive strength, 36% in split tensile strength and 50% in flexural strength. For strength characteristics, the results showed a gradual decrease in compressive strength, split tensile strength, flexural strength and modulus of elasticity as the percentage of recycled aggregate used in the specimens increased. 100% replacement of natural coarse aggregate by recycled aggregate resulted in 43% savings in the cost of coarse aggregates and 9% savings in the cost of concrete.

With the sustainable development of the construction industry, recycled aggregate (RA) has been widely used in concrete preparation to reduce the environmental impact of construction waste. Compressive strength is an essential measure of the performance of recycled aggregate concrete (RAC). In order to understand the correspondence between relevant factors and the compressive strength of recycled concrete and accurately predict the compressive strength of RAC, this paper establishes a model for predicting the compressive strength of RAC using machine learning and hyperparameter optimization techniques. RAC experimental data from published literature as the dataset, extreme gradient boosting (XGBoost), random forest (RF), K-nearest neighbour (KNN), support vector machine regression Support Vector Regression (SVR), and gradient boosted decision tree (GBDT) RAC compressive strength prediction models were developed. The models were validated and compared using correlation coefficients (R2), Root Mean Square Error (RMSE), mean absolute error (MAE), and the gap between the experimental results of the predicted outcomes. In particular, The effects of different hyperparameter optimization techniques (Grid search, Random search, Bayesian optimization-Tree-structured Parzen Estimator, Bayesian optimization- Gaussian Process Regression) on model prediction efficiency and prediction accuracy were investigated. The results show that the optimal combination of hyperparameters can be searched in the shortest time using the Bayesian optimization algorithm based on TPE (Tree-structured Parzen Estimator); the BO-TPE-GBDT RAC compressive strength prediction model has higher prediction accuracy and generalisation ability. This high-performance compressive strength prediction model provides a basis for RAC’s research and practice and a new way to predict the performance of RAC.

The modern world is witnessing the development of extremely demanding structures right now. Concrete is an extremely important and widely used material. However, the reuse of construction waste is crucial when considering life cycle assessment (LCA) and the efficient recycling of construction materials. Given its potential to address sustainability and environmental challenges, using recycled materials in construction has attracted much attention. This study aims to examine what happens to concrete's compressive strength when recycled coarse aggregates (RCA) replace some of the conventional coarse aggregates. The aim is to study the impact of RCA content on the mechanical characteristics of concrete and determine its appropriateness for use in structural applications. The main objectives of this work is to determine and compare the Density, Workability, Compressive strength, split tensile strength, and Flexural strength of Recycled aggregate in relation to conventional aggregate also to ascertain the ideal percentage of aggregate replacement with recycled aggregate and to calculate the percentage change in strength for recycled aggregate at different levels of replacement. The research methodology involves carrying out laboratory experiments to examine the compressive strength, split tensile strength, and flexural strength of concrete specimens with different proportions of recycled concrete aggregate (RCA) substitutions. The typical coarse aggregates are substituted by recycled aggregates in various proportions (5%, 10%, 15%, 20%, 25%, and 30%) by weight in M25 grade concrete. The formulation of the concrete mix design is meticulously done to provide a consistent water-cement ratio and other crucial criteria. The study involves the process of casting and curing concrete specimens that contain varying amounts of recycled concrete aggregate (RCA). At certain curing ages (e.g., 7 days and 28 days), the specimens are tested for compressive strength, split tensile strength, and flexural strength in order to assess their mechanical capabilities. After that, the outcomes are contrasted with control samples made entirely of conventional coarse aggregates. Primary finding shows workability declines as the proportion of recycled trash rises, but it is kept stable with mixing.

An experimental investigation to predict the compressive strength of lightweight Ceramsite aggregate UHPC using boosting and bagging techniques

The incorporation of recycled concrete aggregates (RCAs) into self-compacting concrete (SCC) represents a critical sustainable construction strategy addressing both construction waste management and natural resource conservation. However, predicting the compressive strength of recycled aggregate self-compacting concrete (RASCC) remains challenging due to complex nonlinear interactions among mixture parameters. This study develops a robust predictive framework using ensemble machine learning algorithms to accurately estimate RASCC compressive strength across diverse mixture compositions. A comprehensive database comprising 301 experimental specimens with 18 input variables—including curing age, binder components, water-to-binder ratio, recycled aggregate properties, and supplementary cementitious materials—was systematically analyzed. Four advanced modeling approaches were evaluated: Light Gradient Boosting Machine (LightGBM), Categorical Boosting (CatBoost), Stacked Generalization with Ridge regression meta-learner, and Voting ensemble with Non-Negative Least Squares optimization. The Stacking ensemble model demonstrated superior predictive performance on the independent test set, with R2 = 0.963, RMSE = 3.321 MPa, and MAE = 2.506 MPa. Rigorous residual analysis confirmed model validity through satisfaction of normality, homoscedasticity, and independence assumptions. SHAP interpretability analysis identified specimen age as the dominant predictor, followed by recycled aggregate density and water-to-binder ratio, while elucidating the complex nonlinear contributions of supplementary cementitious materials including fly ash and ground granulated blast furnace slag. The developed framework demonstrates practical applicability for predicting RASCC compressive strength across conventional to high-performance grades, facilitating sustainable mix design optimization while maintaining structural performance requirements, and advancing circular economy principles through confident integration of recycled aggregates in SCC applications.

Green recycled aggregate concrete (RAC) with high strength and low shrinkage is prepared based on recycled coarse aggregate produced by the particle-shaping and aggregate-strengthening method and green low-carbon new cement. This not only effectively alleviates the shortage of natural resources, but also improves the performance of recycled aggregate concrete, which is of great significance for multi-channel resource utilization of construction waste. In this study, three kinds of recycled coarse aggregates (RCA), including simple crushing recycled coarse aggregate (JD-RCA), one-time particle-shaping recycled coarse aggregate (KL-RCA) and two-time particle-shaping recycled coarse aggregate (EKL-RCA), were prepared from the preparation technology of recycled aggregate, and high belite sulphoaluminate cement with excellent performance was used. The effects of aggregate quality, aggregate replacement ratio, and cementitious material content on mechanical properties and shrinkage properties of green recycled aggregate concrete were studied in comparison with ordinary Portland cement-based recycled aggregate concrete. The testing results show that the particle-shaping method can effectively improve the aggregate quality. The compressive strength and dry shrinkage performance of recycled aggregate concrete made of particle-shaped aggregate are only a little different from those of natural aggregate concrete, and even the performance of recycled aggregate is better than that of natural aggregate concrete under the condition of a low replacement ratio of recycled aggregate. In addition, high belite sulphoaluminate cement-based recycled aggregate concrete (HBRAC) not only has early strength and rapid hardening, but also has excellent drying shrinkage resistance, and its shrinkage rate can be reduced to more than 75% compared with ordinary Portland cement-based recycled aggregate concrete (OPRAC).

Recycled geopolymer aggregates as coarse aggregates for Portland cement concrete and geopolymer concrete: Effects on mechanical properties

Objective: The objective is to determine the effect of treated RCA on the mechanical strength of hardened concrete, and determine the influence of treated RCA on water transport within the matrix of hardened concrete. Aim: The study aims to use pozzolanic concentration of cement and RHA slurry to pretreat coarse RCA for full replacement of NCA for sustainable concrete production. Background: One of the common modes of disposal of construction waste is strained landfills. This study proposes a constructive way of recovering the coarse aggregates of recycled aggregates from laboratory concrete waste taken through a focused treatment process with pozzolan. The study aims to replace the proportions of the cement slurry to treat and replace with rice husk ash (RHA) for the modification of recycled concrete aggregate (RCA) and replace fully natural coarse aggregate (NCA) for sustainable concrete production. One of the main challenges of recycled concrete aggregate (RCA) is the fractured surface, which often results in reduced strength and loss of concrete durability. The strength of concrete with RCA is principally influenced by the fractured hardened paste attached to the surface, predominantly characterized by voids and gaps resulting from its production using a mechanical method. High water penetration within and around the aggregate matrix due to the presence of micro- and macro-cracks results in strength and durability deterioration of concrete. Methods: This study addresses these problems by specifying aggregate treatment with a pozzolanic slurry of cement and RHA. This was conducted in two (2) phases; first, to determine the optimum proportion of cement slurry concentration (0, 20, 40, 60, 80, and 100%), and second, to increase the silicate formation in cement slurry by substituting the proportions of cement with RHA (0, 5, 10, and 15%). The performance of concrete was evaluated by mechanical strength (compressive and split tensile strength) at 7, 28, and 56 days and water transport within the concrete (surface suction and submergence) at 28 and 56 days of curing age. Results: The use of cement slurry treatment at various concentrations generally increased compressive strength at 28 days compared to untreated RCA. The findings show that at a 20% concentration of cement slurry, the increase in compressive strength is 38.4%. Upon replacement of cement in treatment slurry of RCA with 5% RHA, the increase in compressive strength was 61%, while the split tensile strength increased by 10.2%. The amount of water absorbed was observed after 30 minutes of full submergence in water, indicating that samples with 20% cement concentration at 28 days resulted in the lowest water absorption of 2.01%. However, with the replacement of 5% RHA, the water absorption slightly increased to 2.3%, but still less than the requirement of 3.5%. The corresponding test results for water suction by capillarity, initial and secondary sorptivity coefficients (Si and Ss) were found to be 0.0330mm/√sec. (1.98mm/√hr.) and 7.95×10−4mm/√sec. (0.04mm/√hr.) respectively, all less than 6mm/√hr requirement. This improved performance was attributed to the increased strength and structure of the interfacial transition zone (ITZ) around the RCA. Conclusion: Based on the results of mechanical strength and water transport, characteristics of the new concrete were improved when coarse RCAs were pretreated with 20% pozzolanic concentration containing cement and RHA. The use of a pozzolanic treatment slurry containing 15% cement and 5% RHA to treat coarse RCA can produce concrete with characteristics similar to those of natural concrete aggregates. This study presents a methodological utilization and improvement of RCA wastes for field application.

Experimental study of the flexural implementation of high-strength reinforced recycled aggregate mixture is investigated for 100 × 150 × 1000 mm beams of reinforced concrete with different coarse recycled aggregate ratios plus three steel reinforcement ratio (ρ) casting and testing at 28 days. The aim of this research is to use coarse recycled aggregate ratios as a natural coarse aggregate replacement by (0, 10, 20, 30, 40, 50, 60, 70, 80, 90, and 100%) in high strength concrete. Former investigations experiments have been carried out and provided on the characteristics of the mixture prepared with recycled aggregates, such as strength in compressive, workability, flexural tensile strength, and splitting tensile strength. Prism and cylinders have been prepared, and specimens have been tested at ages 7, 14, 28 days of curing. This investigation considers the flexural efficiency of reinforced recycled aggregate mixture beams in describe as cracking and ultimate flexural strength, and load-deflection correlation, the moment at mid-span, and ductility and failure modes. Results show that 100% recycled coarse aggregate degrades the concrete strength relevant to 18.0, 18.9, and 19.3%, 16.2, 16.6, and 19.4% and 16.0, 16.2, and 18.2% in expression of compressive, splitting and flexural strength with tensile for 7, 14 and 28 days, reduces the workability by 33.3% and decreases the cracking and ultimate load with the improved rate of recycled coarse aggregate regarding of 39.4 and 45.1% and 10.9% for ductility for the beams with ratio of reinforcement of ρ=0.0157%. Decreasing the reinforcement ratios with ρ=0.0105 and 0.0067% while keeping the strength of concrete constant 100% recycled coarse aggregate decreases the cracking and ultimate load of about 11.3 and 20.8% and 15.6 and 24.4% respectively and 2.0 and 4.3% for ductility.

The usage of recycled aggregate (RA) for concrete works is spreading for gaining wide acceptance around the world, and total amount of RA utilized in concrete industry increases day-by-day. RA is used in many concrete products, such as concrete interlocking blocks, pavements, etc. One of the physical causes of concrete deterioration is surface wear or loss of mass for concrete products which are especially exposed to vehicular traffic (i.e., industrial floors and pavements). Besides, the loss of mass by concrete surface wear becomes the primary factor of one or more processes of chemical deterioration as a result of an increase in permeability due to mass loss of the concrete surface wear. The main goal of this research is to investigate the abrasion behavior of recycled aggregate concrete (RAC) with/without silica fume (SF) related with the mechanical and the physical properties of RAC. In this research, lower quality RA (<15 MPa) is utilized as recycled coarse aggregate (RCA1 and RCA2) and recycled fine aggregate (RFA) with 25, 25, and 50% replacement ratios, respectively, in concrete mixes. As a result, RCA2 with/without SF gives better performance than RCA1 and RFA if abrasion resistance is considered only. Besides, RFA utilization in concrete decreases the abrasion performance of RAC. Also the relations of compressive strength-abrasion loss and elasticity modulus-abrasion loss are examined, and high regression coefficient (R) for the relations is found. Keywords: abrasion resistance, mechanical properties, physical properties, recycled aggregate, silica fume

The long term properties of Concrete containing Recycled Coarse Aggregates (CRCA) were investigated. Initially, series of short term tests were carried out to determine the density, workability, absorption and compressive strength of concrete with and without locally available recycled aggregates from demolition wastes. The water cement ratio of 0.52 was adopted for all the mixes, while the coarse aggregate in concrete was replaced with 0%, 25%, 50% and 100% recycled coarse aggregates. The test results indicated that the replacement of normal coarse aggregates by recycled aggregates up to 25% had no significant effect on the compressive strength but higher levels of replacement reduced the compressive strength. A replacement level of 100% caused a reduction of about 27% in compressive strength. The shrinkage deformation characteristics of concrete made with 25% recycled coarse aggregate were compared with those of normal concrete. Eight 10cm x 10cm x 40cm concrete sealed and unsealed short columns made with and without RCA, were under investigation for 120 days. The results of the investigation showed that the basic shrinkage strains of normal concrete is about 1.07 times greater than that of CRCA (with Original Research Article Article British Journal of Applied Science & Technology, 4(12): 1791-1807, 2014 1792 25% RCA). The drying and total shrinkage strains of CRCA (with 25% RCA) are respectively 2.56 and 1.26 times greater than that of normal concrete. 25% RCA content in concrete was therefore considered for use in load bearing structural elements.

Modeling of necking area reduction of carbon steel in hydrogen environment using machine learning approach

The surface zone of the recycled aggregate plays a significant role in the strength and durability of concrete. In this research, various concrete series incorporating natural coarse aggregate [NCA], untreated recycled coarse aggregate RCA(UN), mechanical scrubbing treated coarse aggregate RCA(MS), sulfuric acid treated coarse aggregate RCA(H2SO4) and hydrochloric acid treated aggregate RCA(HCl) were investigated. Manufactured sand was used as an alternative to natural river sand in five series of concrete with 100% replacement ratio. The mechanical properties: compressive, split tensile, and flexural strength, respectively, were determined at the age of 28 and 56 days. Durability properties: water absorption, drying, sulfuric acid, sodium sulphate and sorptivity test were determined at the age of 28, 56 and 90 days and the results were compared to reference concrete. According to the results, RCA(UN) was found to marginally decreased the concrete performance, and the above mentioned treated aggregate shows a significant improvement in overall performance compared to RCA(UN), especially RCA(MS), which enhances the concrete more significantly than other methods. Moreover, the scanning electron microscope results also emphasize the enhanced microstructure of treated recycled aggregate in concrete. The photographic view of the cross sections of the samples shows a superior interfacial transition zone in all kinds of treated aggregate concrete than in untreated recycled aggregate concrete.

Evaluating compressive strength of concrete made with recycled concrete aggregates using machine learning approach

This research targets at providing insights on how coarse aggregate prepared from recycled concrete differs from normal coarse aggregate. The recycled coarse aggregate in this context is considered only if it comes from demolished concrete waste; construction waste has been kept out of the scope of the study. For this research, the recycled coarse aggregate was collected from Pasakha industrial dump yard and was manually crushed to the required size. The study aims at checking the feasibility of using recycled aggregate in concrete making. Aggregate tests were first conducted in the college laboratory to check whether the recycled aggregate met the basic strength and quality requirements as per standards. Tests on natural (normal) coarse aggregate were also conducted as reference or control tests. After obtaining satisfactory results from the aggregate tests, the study proceeded to test concrete made from both recycled coarse aggregate and normal aggregate using the same cement and fine aggregate. Workability test, shrinkage test, and compressive strength test were conducted. A total of 46 specimens were prepared — 40 for compressive strength testing and 6 for shrinkage testing. The paper also presents the stress-strain behavior and elastic modulus of both types of concrete samples.