Damage assessment and diagnosis of hydraulic concrete structures using optimization-based machine learning technology

Article Engineering, Mechanical

An efficient stochastic-based coupled model for damage identification in plate structures

Long Viet Ho et al.

Summary: The study introduces a damage identification method based on mode shape derivatives and utilizes ALOANN strategy to improve performance in localizing damage in plate structures.

ENGINEERING FAILURE ANALYSIS (2022)

添加到收藏夹

Article Engineering, Civil

Performance-improved TSVR-based DHM model of super high arch dams using measured air temperature

Dongyang Yuan et al.

Summary: A new measured air temperature-based displacement health monitoring model for super high arch dams is proposed in this study, utilizing twin support vector regression and optimized using grey wolf optimizer algorithm. Experimental results demonstrate that the new model can better capture the thermal displacement variations, significantly improving predictive accuracy and showing excellent long-term predictive capability.

ENGINEERING STRUCTURES (2022)

添加到收藏夹

Article Construction & Building Technology

Assessment of damage in hydraulic concrete by gray wolf optimization-support vector machine model and hierarchical clustering analysis of acoustic emission

Xing Li et al.

Summary: In this study, the fracture process of hydraulic concrete was analyzed using acoustic emission data, revealing three distinct stages of fracture and characterizing different failure modes through specific parameters. Hierarchical clustering analysis and a support vector machine model successfully quantified the degree of damage in the concrete. Therefore, combining acoustic emission and machine learning offers an effective method for damage assessment.

STRUCTURAL CONTROL & HEALTH MONITORING (2022)

添加到收藏夹

Article Engineering, Mechanical

A hybrid PSO and Grey Wolf Optimization algorithm for static and dynamic crack identification

Faisal Al Thobiani et al.

Summary: This paper presents an inverse problem for crack identification in two-dimensional structures using XFEM, GWO, and IGWO. Experimental results show that IGWO outperforms GWO in terms of accuracy. Additionally, GWO and IGWO are employed to improve ANN parameters for crack length prediction.

THEORETICAL AND APPLIED FRACTURE MECHANICS (2022)

添加到收藏夹

Article Engineering, Civil

Deep convolutional neural network for multi-level non-invasive tunnel lining assessment

Bernardino Chiaia et al.

Summary: In recent years, there has been increasing emphasis on the development of automated procedures for infrastructure control. This paper introduces a multi-level strategy that utilizes convolutional neural networks to identify and classify potential critical situations. By analyzing various structural phenomena, this strategy allows for accurate identification of structural conditions, thereby reducing the time and costs associated with testing and inspections.

FRONTIERS OF STRUCTURAL AND CIVIL ENGINEERING (2022)

添加到收藏夹

Article Mathematics

Pearson Correlation and Discrete Wavelet Transform for Crack Identification in Steel Beams

Morteza Saadatmorad et al.

Summary: A novel approach is introduced in this paper for crack identification of beam structures, which combines statistical characteristics and discrete wavelet transforms to improve the accuracy of damage detection.

MATHEMATICS (2022)

添加到收藏夹

Article Engineering, Civil

Data-driven crack behavior anomaly identification method for concrete dams in long-term service using offline and online change point detection

Yangtao Li et al.

Summary: The study emphasizes the importance of accurately assessing the impact of cracks on the structural safety of concrete dams and introduces various offline and online changepoint detection methods. The experimental results demonstrate that the proposed model can accurately and timely detect the change points in crack behavior and obtain exact dates of change.

JOURNAL OF CIVIL STRUCTURAL HEALTH MONITORING (2021)

添加到收藏夹

Article Engineering, Civil

Detection of damage locations and damage steps in pile foundations using acoustic emissions with deep learning technology

Alipujiang Jierula et al.

Summary: This study proposes a new method for determining damage locations in pile foundations using deep learning and acoustic emission data, as well as for classifying damage step conditions. The study explores the impact of various acoustic emission parameters, sensor numbers and locations, and time difference on the accuracy of damage location prediction. The new method aims to continuously detect and evaluate the damage of pile foundations in service.

FRONTIERS OF STRUCTURAL AND CIVIL ENGINEERING (2021)

添加到收藏夹

Article Engineering, Civil

Automated classification of civil structure defects based on convolutional neural network

Pierclaudio Savino et al.

Summary: This paper presents a method for automating concrete damage classification using a deep convolutional neural network, which was designed after an experimental investigation and transfer-learning technique. By collecting different image configurations from real-world situations, the network's robustness was increased, and the most suitable network for concrete damage classification was found.

FRONTIERS OF STRUCTURAL AND CIVIL ENGINEERING (2021)

添加到收藏夹

Article Computer Science, Interdisciplinary Applications

A hybrid computational intelligence approach for structural damage detection using marine predator algorithm and feedforward neural networks

Long Viet Ho et al.

Summary: FE-based SHM algorithms aim to update damage indices by solving an optimization problem to minimize the difference between real structure response and FE model response. The study proposes an effective approach using MPA for damage detection, with superior performance compared to other optimization algorithms. The MPAFNN method shows promising results for both classification and damage identification tasks.

COMPUTERS & STRUCTURES (2021)

添加到收藏夹

Article Construction & Building Technology