International Journal of Fatigue · 2024 · 10 citations · 58 references
ComputingMaterials EngineeringMaterials ScienceBuilding MaterialsComputational Materials ScienceMachine LearningConcrete ConstructionConcrete MaterialsDeep LearningEnsemble LearningFatigueMaterial PropertiesEnsemble-based PredictionFatigue LifeFatigue Life PredictionFatigue ResistanceFlexural Fatigue LifeGrey Wolf OptimizerMachine Learning In ConstructionReinforced Concrete
The incorporation of fibers into concrete enhances its fatigue resistance while also increasing the variability in flexural fatigue life , necessitating the development of advanced predictive models. To address this challenge, this study innovates by developing a stacking ensemble prediction model aimed at accurately predicting the flexural fatigue life of fiber-reinforced concrete (FRC). This model integrates the Deep Autoencoder Network, XGBoost, Random Forest, and Deep Neural Networks optimized by the Grey Wolf Optimizer algorithm. Initially, a dataset on the flexural fatigue of FRC was meticulously established. Subsequently, the model was rigorously evaluated using this dataset, with the results demonstrating its effectiveness in accurately predicting the fatigue life of FRC . Furthermore, SHAP analysis was utilized to interpret the relationship between input features and the fatigue life of FRC. In essence, this research offers a comprehensive and flexible predictive framework for the fatigue life of FRC, enhancing the comprehension and practical utilization of this material in construction and engineering projects, and presenting a promising avenue for future advancements in materials science and engineering.
58
Leo Breiman · Machine Learning · 2001 · 126.2K citations · Full text
Seyedali Mirjalili, Seyed Mohammad Mirjalili, Seyed Mohammad Mirjalili et al. · Advances in Engineering Software · 2014 · 18.4K citations · Full text