Machine Learning-Driven Predictive Modeling for FRP Strengthened Structural Elements: A Review of AI-Based Damage Detection, Fatigue Prediction, and Structural Health Monitoring DOI

James Avevor,

Michael Adeniyi,

Lawrewnce Anebi Enyejo

и другие.

Опубликована: Авг. 28, 2024

The integration of Machine Learning (ML)-driven predictive modeling has revolutionized the assessment and optimization Fiber-Reinforced Polymer (FRP) strengthened structural elements, offering advanced methodologies for damage detection, fatigue prediction, health monitoring (SHM). This review provides a comprehensive analysis AI-based techniques, including deep learning (DL), convolutional neural networks (CNNs), recurrent (RNNs), support vector machines (SVMs), ensemble methods, evaluating mechanical performance longevity FRP-reinforced structures. study explores how ML algorithms process sensor-acquired data from acoustic emission (AE), digital image correlation (DIC), fiber Bragg gratings (FBGs), vibration measurements to predict crack initiation, failure, progressive degradation in composite-strengthened bridges, high-rise buildings, aerospace Additionally, this investigates thermomechanical aeroelastic effects on FRP-strengthened elements under dynamic loading conditions, highlighting ability ML-based hybrid models enhance accuracy multi-variable stress-strain behavior prediction. incorporation physics-informed (PINNs) AI-physics further refines localization severity estimation, addressing uncertainties material anisotropy, bond degradation, environmental aging effects. Moreover, advances transfer federated (FL) enable real-time SHM large-scale infrastructure by leveraging cloud-based edge computing frameworks decentralized anomaly detection maintenance.This paper also discusses twin (DT) technology with SHM, enabling simulation, life-cycle Challenges such as model interpretability, scarcity, computational efficiency are examined, along potential explainable AI (XAI), uncertainty quantification (UQ), reinforcement (RL) optimizing decision-making processes sustainability. concludes identifying future research directions methodologies, adaptive frameworks, quantum-enhanced modeling, aiming resilience durability systems civil engineering applications.

Язык: Английский

Interpretable Machine Learning Insights into Wildfire Damage Drivers in California, USA DOI Creative Commons
Yiming Jia, Eyitayo A. Opabola

International Journal of Disaster Risk Reduction, Год журнала: 2025, Номер unknown, С. 105610 - 105610

Опубликована: Май 1, 2025

Язык: Английский

Процитировано

0
Machine Learning-Driven Predictive Modeling for FRP Strengthened Structural Elements: A Review of AI-Based Damage Detection, Fatigue Prediction, and Structural Health Monitoring DOI

James Avevor,

Michael Adeniyi,

Lawrewnce Anebi Enyejo

и другие.

Опубликована: Авг. 28, 2024

The integration of Machine Learning (ML)-driven predictive modeling has revolutionized the assessment and optimization Fiber-Reinforced Polymer (FRP) strengthened structural elements, offering advanced methodologies for damage detection, fatigue prediction, health monitoring (SHM). This review provides a comprehensive analysis AI-based techniques, including deep learning (DL), convolutional neural networks (CNNs), recurrent (RNNs), support vector machines (SVMs), ensemble methods, evaluating mechanical performance longevity FRP-reinforced structures. study explores how ML algorithms process sensor-acquired data from acoustic emission (AE), digital image correlation (DIC), fiber Bragg gratings (FBGs), vibration measurements to predict crack initiation, failure, progressive degradation in composite-strengthened bridges, high-rise buildings, aerospace Additionally, this investigates thermomechanical aeroelastic effects on FRP-strengthened elements under dynamic loading conditions, highlighting ability ML-based hybrid models enhance accuracy multi-variable stress-strain behavior prediction. incorporation physics-informed (PINNs) AI-physics further refines localization severity estimation, addressing uncertainties material anisotropy, bond degradation, environmental aging effects. Moreover, advances transfer federated (FL) enable real-time SHM large-scale infrastructure by leveraging cloud-based edge computing frameworks decentralized anomaly detection maintenance.This paper also discusses twin (DT) technology with SHM, enabling simulation, life-cycle Challenges such as model interpretability, scarcity, computational efficiency are examined, along potential explainable AI (XAI), uncertainty quantification (UQ), reinforcement (RL) optimizing decision-making processes sustainability. concludes identifying future research directions methodologies, adaptive frameworks, quantum-enhanced modeling, aiming resilience durability systems civil engineering applications.

Язык: Английский

Процитировано

0