Optimizing high-strength concrete compressive strength with explainable machine learning

Multiscale and Multidisciplinary Modeling Experiments and Design, Год журнала: 2025, Номер 8(3)

Опубликована: Фев. 3, 2025

Язык: Английский

---

Estimating the compressive and tensile strength of basalt fibre reinforced concrete using advanced hybrid machine learning models

Structures, Год журнала: 2025, Номер 71, С. 108138 - 108138

Опубликована: Янв. 1, 2025

Язык: Английский

---

AI-driven Modeling for the Optimization of Concrete Strength for Low-Cost Business Production in the USA Construction Industry

Engineering Technology & Applied Science Research, Год журнала: 2025, Номер 15(1), С. 20529 - 20537

Опубликована: Фев. 2, 2025

The need to develop ecologically friendly sustainable building materials is made apparent by the worldwide construction industry's substantial contribution global greenhouse gas emissions. use of supplemental in concrete one potential solution lessen environmental footprint. Thus, purpose this work Machine Learning (ML) algorithms forecast and create an empirical formula for Compressive Strength (CS) with materials. Six distinct ML models—XGBoost, Linear Regression, Decision Tree, k-Nearest Neighbors, Bagging, Adaptive Boosting—were trained tested using a dataset that included 359 experimental data varying mix proportions. most significant factors used as input parameters are cement, aggregates, water, superplasticizer, silica fume, ambient curing, material. Several statistical measures, such Mean Absolute Error (MAE), coefficient determination (R2), Square (MSE), were evaluate models. XGBoost model outperformed other models R2 values 0.99 at training stage. To ascertain how affected outcome, feature importance analysis Shapely Additive exPlanations (SHAP) was conducted. It demonstrated curing age cement type significantly strength high SHAP values. By eliminating procedures, reducing demand labor resources, increasing time efficiency, offering insightful information enhancing manufacturing concrete, research advances low-cost production USA industry.

Язык: Английский

---

Investigating the effect of locally available volcanic ash on mechanical and microstructure properties of concrete

REVIEWS ON ADVANCED MATERIALS SCIENCE, Год журнала: 2025, Номер 64(1)

Опубликована: Янв. 1, 2025

Abstract Environmental degradation is developing due to rising pollution from the depletion of raw materials and growing mandate for concrete goods. Investigators experts have focused on creating sustainable utilizing renewable elements. Volcanic ash (VA) a promising supplementary cementitious material among these minerals. Therefore, it crucial examine attributes voids in aggregate how they impact performance concrete. VA Gini Chilas (Gigilat Baltistan) was used prepare specimens. Mixing regimes with altering concentrations ranging 0 40% replacement cast. Water-to-cement ratio reserved persistent all mixes. Chemical compositions properties relation workability, density, compressive strength were carried out. In addition, thermo-gravimetric analysis, scanning electron microscope (SEM), X-ray diffraction analysis also examined. The results reveals that 10% gives an adamant response. This natural pozzolanic effect details creation additional dense gel (C–S–H), deviation cracks observed SEM. 10 exhibits thermally stable behavior at temperature less percentage mass loss. However, up can exhibit better than normal

Язык: Английский

---

Enhancing the Predictive Accuracy of Marshall Design Tests Using Generative Adversarial Networks and Advanced Machine Learning Techniques

Materials Today Communications, Год журнала: 2025, Номер unknown, С. 112379 - 112379

Опубликована: Март 1, 2025

Язык: Английский

---

Machine Learning-Driven Flexural Performance Prediction and Experimental Investigation of Glass Fiber-Reinforced Polymer Bar-Reinforced Concrete Beams

Polymers, Год журнала: 2025, Номер 17(6), С. 713 - 713

Опубликована: Март 7, 2025

This study experimentally examines the flexural performance, crack formation patterns, and failure mechanisms of glass fiber-reinforced polymer (GFRP) bar-reinforced concrete beams with varying compressive strengths (low, moderate, high), addressing a gap in current literature. Furthermore, it employs an innovative machine learning approach to enhance analysis. Nine RC reinforced GFRP bars, having low (CC20), moderate (CC30), high (CC40), each measuring 150 × 200 1100 mm, were fabricated tested under three-point bending conditions. Through integration tests learning-based prediction models, this connects experimental findings advanced analytical approaches. One key innovations is use eighteen ML regression models implemented Python’s PyCaret library, achieving impressive average accuracy 91.5% for beam deflection values. In particular, Ada Boost Regressor Gradient Boosting performed exceptionally well on beams, providing highest number consistent highly accurate predictions, making them very useful tools ultimate load-carrying capacity/deflection predictions. The outcomes identified clear mechanisms: CC20, CC30, CC40 typically developed single, large at midpoint. Although capacity bar improved higher strength, CC20 CC30 displayed more ductile behavior than beams. was determined be approximately 74% that Regardless strength class, absence shear cracks prevention sudden are considered major advantages using reinforcement.

Язык: Английский

---

Machine Learning-Based Partition Method for Cyclic Development Mode of Submarine Soil Martials from Offshore Wind Farms

Journal of Marine Science and Engineering, Год журнала: 2025, Номер 13(3), С. 533 - 533

Опубликована: Март 10, 2025

Offshore wind turbines are subjected to long-term cyclic loads, and the seabed materials surrounding foundation susceptible failure, which affects safe construction normal operation of offshore turbines. The existing studies mechanical properties submarine soils focus on accumulation strain liquefaction, few targeted conducted hysteresis loop under loads. Therefore, 78 representative soil samples from four farms tested in study, behaviors different confining pressures CSR investigated. experiments reveal two unique development modes specify critical five martials testing conductions. Based dynamic triaxial test results, machine learning-based partition models for mode were established, discrimination accuracy discussed. This study found that RF model has a better generalization ability higher than GBDT discriminating soil, achieved prediction 0.96 recall 0.95 dataset, provides an important theoretical basis technical support design

Язык: Английский

---

Experimental and Numerical Investigation on the Effect of Different Types of Synthetic Fibers on the Flexure Behavior and Mechanical Properties of 3D Cementitious Composite Printing Provided with Cement CEM II/A-P

Buildings, Год журнала: 2025, Номер 15(7), С. 1201 - 1201

Опубликована: Апрель 6, 2025

Concrete printing in three dimensions is believed to be an innovative construction method. Numerous researchers conducted laboratory experiments over the past decade examine behavior of concrete mixtures and material properties that are pertinent 3D industry. Furthermore, global warming effect being further exacerbated by increased use cement, which increases carbon dioxide (CO2) emissions pollution. Various standards endorse utilization Portland-composite cement mitigate CO2 emissions, particularly CEM II/A-P. This research provides experimental numerical study evolution cementitious composite utilizing II/A-P for three-dimensional printing, combining different types synthetic fiber. The thorough analysis includes combinations integrating diverse fiber (polypropylene, high-modulus polyacrylonitrile, alkali-resistant glass fibers) alongside a reference mixture devoid distinct high modulus were evaluated assess their impact on (i) flowability mortar slump flow test fresh concrete, (ii) compressive strength, (iii) uniaxial tensile (iv) splitting (v) flexural strength. Previous designed cylinder stability determine shape layers capacity support stresses from subsequent layers. corroborated findings with finite element software ANSYS 2023 R2. performance examined beams was validated using Menetrey–Willam constitutive model, has recently been incorporated into ANSYS. data indicated incorporation enhanced concrete’s combination including fibers. results demonstrated efficacy featuring linear softening yield function accurately simulating analyzed various types.

Язык: Английский

---

A scientometric review on the utilization of copper slag as a substitute constituent of ordinary Portland cement concrete

REVIEWS ON ADVANCED MATERIALS SCIENCE, Год журнала: 2025, Номер 64(1)

Опубликована: Янв. 1, 2025

Abstract This article presents a scientometric review on the utilization of copper slag (CS) as substitute constituent in ordinary Portland cement concrete, with greater focus analyzing CS supplementary cementitious material (SCM). The was conducted through comprehensive analysis literature using Scopus and VOSviewer, examining publication trends, document types, subject areas, leading contributors, overall progression research concrete. revealed substantial increase publications between 2015 2022, journal “Construction Building Materials” country “India” identified most influential field. methodology involved filtering relevant documents to impactful research, which then critically analyzed assess fresh hardened properties findings indicate that incorporating 5–10% an SCM can significantly enhance mechanical durability also found improve concrete by imparting micro-filler effect, thereby densifying structure. Additionally, contributes ecological benefits heavy metals into matrix, preventing their leaching, aiding environmental conservation. Despite these promising results, acknowledges long-term performance remains critical area needs further investigation.

Язык: Английский

---

Developing robust structure: Multi‐expression programming for anticipating mechanical properties of shape memory alloy‐confined concrete cylinders

Structural Concrete, Год журнала: 2025, Номер unknown

Опубликована: Апрель 24, 2025

Abstract This study emphasizes the expansion of novel relationships to determine maximum compressive stress, corresponding strain, ultimate and strain in order enhance precision practicality predicting behavior SMA‐confined concrete (SMACC) with spirals. It develops predictive equations for mechanical properties SMACC cylinders using multi‐expression programming (MEP) method. The MEPX software is employed derive optimal by collecting experimental data from 42 cylindrical specimens subjected uniaxial compression confined SMA findings show that developed MEP‐based not only provide practical equations, but also produce more precise results.

Язык: Английский

---