Prediction and optimization model of sustainable concrete properties using machine learning, deep learning and swarm intelligence: A review

Journal of Building Engineering, Год журнала: 2023, Номер 80, С. 108065 - 108065

Опубликована: Ноя. 3, 2023

Язык: Английский

---

Predictive models in machine learning for strength and life cycle assessment of concrete structures

Automation in Construction, Год журнала: 2024, Номер 162, С. 105412 - 105412

Опубликована: Апрель 3, 2024

Язык: Английский

---

Numerical and machine learning models for concentrically and eccentrically loaded CFST columns confined with FRP wraps

Structural Concrete, Год журнала: 2024, Номер unknown

Опубликована: Сен. 22, 2024

Abstract Previous research largely concentrated on predicting load‐carrying capacities of concrete‐filled steel tubes (CFST) confined with fiber‐reinforced polymer (FRP) wraps under pure concentric loads, neglecting the more complex failure mechanisms that occur real‐life eccentric loading conditions. This study, therefore, employs both finite element modeling (FEM) and machine learning methods to accurately predict load‐bearing analyzed a comprehensive dataset comprising 128 experimental tests an equivalent number FEM simulations designed evaluate their performance. These models have been thoroughly generated validated against existing literature. Additionally, developed ML models, particularly hybrid deep model, demonstrated significant predictive accuracy, average R 2 value 0.969 across all model folds. Partial dependence analysis further highlighted influence concrete strength interactive effects tube area FRP wrap thickness capacity columns. Furthermore, enhance cost‐efficiency resource management compared traditional laboratory testing, user‐friendly graphical user interface (GUI) has hosted open‐source platform such as GitHub. supports real‐time, precise capabilities promotes collaborative environment for ongoing refinement improvement.

Язык: Английский

---

Machine learning methods for compression capacity prediction and sensitivity analysis of concrete-filled steel tubular columns: State-of-the-art review

Structures, Год журнала: 2025, Номер 72, С. 108259 - 108259

Опубликована: Янв. 16, 2025

Язык: Английский

---

Coupled extreme gradient boosting algorithm with artificial intelligence models for predicting compressive strength of fiber reinforced polymer- confined concrete

Engineering Applications of Artificial Intelligence, Год журнала: 2024, Номер 134, С. 108674 - 108674

Опубликована: Июнь 3, 2024

Язык: Английский

---

Flexural performance of FRP-SWSSC-steel composite beams: Experimental and analytical investigation

Engineering Structures, Год журнала: 2024, Номер 306, С. 117842 - 117842

Опубликована: Март 20, 2024

Язык: Английский

---

Study of the behavior of eccentrically-loaded fiber-reinforced polymer (FRP) confined RC columns using a novel approach based on probabilistic gene expression programming

Mechanics of Advanced Materials and Structures, Год журнала: 2025, Номер unknown, С. 1 - 31

Опубликована: Янв. 7, 2025

Fiber-reinforced polymer (FRP) confining of reinforced concrete columns (RCCs) is a widely adopted strengthening technique in civil engineering. However, accurately predicting the axial strength and ductility eccentrically-loaded FRP-confined RCC remains challenge. In this study, finite element models (FEM) were employed conjunction with novel approach gene expression programming (GEP)-probabilistic based modeling to develop predictive for circular RCCs. Extensive FEM simulations conducted, considering various input parameters such as eccentricity-to-section diameter ratios, height-to-section cover diameter, compressive strength, FRP layer thickness, reinforcement ratios. The study pursues three main objectives: first, assess stress–strain behavior RCCs analyze effects on their lateral strains, ductility; second, mathematical equations using GEP algorithm estimate variables ratios; third, conduct probabilistic evaluation analysis uncertainties predictions resulting from models. database facilitated development data-driven models, achieving correlation coefficients 0.943 0.947 ductility, respectively. External validation sensitivity analyses demonstrated superior performance accuracy proposed compared existing literature Additionally, semi- full-probabilistic was implemented account variables, yielding probability distributions GEP-based ductility. These enable assessment associated risks aid informed decision-making design structural solutions. Overall, developed offer valuable tools purposes contribute advancing understanding optimization RC column under eccentric loading conditions.

Язык: Английский

---

Prediction of impact behavior of notched composite pipes repaired with different composite patches by machine learning algorithms

Journal of Adhesion Science and Technology, Год журнала: 2025, Номер unknown, С. 1 - 38

Опубликована: Фев. 5, 2025

Язык: Английский

---

Empirical modeling of stress concentration factors using finite element analysis and artificial neural networks for the fatigue design of tubular KT‐joints under combined loading

Fatigue & Fracture of Engineering Materials & Structures, Год журнала: 2023, Номер 46(11), С. 4333 - 4349

Опубликована: Авг. 9, 2023

Abstract The hotspot stress (HSS) approach for the fatigue design of tubular joints requires that peak HSS be known. Peak in is usually determined based on concentration factor (SCF) estimated from empirical models developed through extensive experimental investigations and finite element analysis. While occurs at a KT‐joint's crown saddle points, its location may change if joint subjected to combination axial, in‐plane bending, or out‐of‐plane bending loads. This study investigated typical KT‐joint combined loading. Specifically, determine SCF around brace axis have been using analysis artificial neural networks (ANN) simulations. Less than 3% error was noticed between FEA. Hence, ANN‐based equations principle superposition can used calculate rapidly joints. methodology applicable developing other boundary conditions.

Язык: Английский

---

An Artificial Neural Network Model for the Stress Concentration Factors in KT-Joints Subjected to Axial Compressive Load

Materials science forum, Год журнала: 2023, Номер 1103, С. 163 - 175

Опубликована: Окт. 25, 2023

Stress concentration factor (SCF) is usually used to estimate the fatigue life of an offshore joint. Historically, parametric equations were SCF based on a statistical analysis experimental and finite element (FEA) results, reduce cost time. These give at saddle/crown position for simple joints basic load cases. However, modified or defective joints, location maximum can change. In such circumstances, single-point equation cannot be value SCF, as its may have changed from saddle/crown. To our knowledge, there are no general expressions around brace axis accurately. As artificial neural networks (ANN) approximate trend complex phenomena better than conventional data fitting, mathematical model ANN proposed weights biases trained ANN. Nine hundred thirty-seven simulations performed generate training This was empirical SCF. The with less 5% error. current study provides roadmap using FEA modeling in tubular this approach applied any joint type, without design modification damage. Once database similar available, it utilized quickly estimating instead costly experimentation FEA. Optimization further improve accuracy developed model.

Язык: Английский

---