Prediction of the Precast Segment Self-Centering Cfst Bridge Seismic Acceleration Response Based on Temporal Convolutional Networks DOI
Dan Zhang, Guixiang Xue,

Jingli Miao

и другие.

Опубликована: Янв. 1, 2024

Язык: Английский

Comparative Analysis of Machine Learning Models for Predicting the Mechanical Behavior of Bio-Based Cellular Composite Sandwich Structures DOI Open Access
Danial Sheini Dashtgoli, Seyedahmad Taghizadeh,

Lorenzo Macconi

и другие.

Materials, Год журнала: 2024, Номер 17(14), С. 3493 - 3493

Опубликована: Июль 15, 2024

The growing demand for sustainable materials has significantly increased interest in biocomposites, which are made from renewable raw and have excellent mechanical properties. use of machine learning (ML) can improve our understanding their behavior while saving costs time. In this study, the innovative biocomposite sandwich structures under quasi-static out-of-plane compression was investigated using ML algorithms to analyze effects geometric variations on load-bearing capacities. A comprehensive dataset experimental tests focusing loading employed, evaluating three models—generalized regression neural networks (GRNN), extreme (ELM), support vector (SVR). Performance indicators such as R-squared (R2), mean absolute error (MAE), root square (RMSE) were used compare models. It shown that GRNN model with an RMSE 0.0301, MAE 0.0177, R2 0.9999 training dataset, 0.0874, 0.0489, 0.9993 testing set had a higher predictive accuracy. contrast, ELM showed moderate performance, SVR lowest accuracy RMSE, MAE, values 0.5769, 0.3782, 0.9700 training, 0.5980, 0.3976 0.9695 testing, suggesting it limited effectiveness predicting structures. nonlinear load-displacement behavior, including critical peaks fluctuations, effectively captured by both test datasets. progressive improvement performance illustrated, highlighting increasing complexity capability models capturing detailed relationships. superior generalization ability confirmed Taylor diagram Williams plot, majority samples falling within applicability domain, indicating strong new, unseen data. results demonstrate potential advanced accurately predict enabling more efficient cost-effective development optimization processes field materials.

Язык: Английский

Процитировано

8
Eco-friendly approach utilizing banana peel as a renewable additive for Portland cement DOI
Jin Yang,

Zhiliang Dong,

Ying Su

и другие.

Sustainable Chemistry and Pharmacy, Год журнала: 2025, Номер 44, С. 101928 - 101928

Опубликована: Янв. 30, 2025

Язык: Английский

Процитировано

0
Water-resistant and anti-mildew soy protein adhesive with network structures based on reversible boron-oxygen bonds and multiple hydrogen bonds DOI
Siwen Pan,

Dong Ho Kong,

Hui Chen

и другие.

Industrial Crops and Products, Год журнала: 2024, Номер 222, С. 119878 - 119878

Опубликована: Окт. 24, 2024

Язык: Английский

Процитировано

2
Seismic acceleration response prediction method of the PSCFST bridge based on TCN DOI
Guixiang Xue,

Jingli Miao,

Dan Zhang

и другие.

Journal of Constructional Steel Research, Год журнала: 2024, Номер 224, С. 109147 - 109147

Опубликована: Ноя. 10, 2024

Язык: Английский

Процитировано

1
Prediction of the Precast Segment Self-Centering Cfst Bridge Seismic Acceleration Response Based on Temporal Convolutional Networks DOI
Dan Zhang, Guixiang Xue,

Jingli Miao

и другие.

Опубликована: Янв. 1, 2024

Язык: Английский

Процитировано

0