Vibration analysis and identification of breathing cracks in beams subjected to single or multiple moving mass using online sequential extreme learning machine

Kourehli, Seyed Sina; Ghadimi, Siamak; Ghadimi, Reza

doi:10.1080/17415977.2018.1479407

Cited by 9 publications

(6 citation statements)

References 18 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Ali and Cha [22] suggested a technique based on deep learning for identifying subsurface damage to steel members in steel truss bridges; the field test results on a steel bridge demonstrated the high accuracy and practicality of the approach. Kourehli and Ghadimi [23] used the emergency learning algorithm of an inline sequential limit learning machine to anticipate the fracture depth and position of a Timoshenko beam. Teng et al [24] used FE numerical analysis and experimental data to provide a considerable amount of damage data for a CNN.…”

Section: Introductionmentioning

confidence: 99%

Deep Learning Damage Identification Method for Steel- Frame Bracing Structures Using Time–frequency Analysis and Convolutional Neural Networks

2023

View full text Add to dashboard Cite

Lattice bracing, commonly used in steel construction systems, is vulnerable to damage and failure when subjected to horizontal seismic pressure. To identify damage, manual examination is the conventional method applied. However, this approach is time-consuming and typically unable to detect damage in its early stage. Determining the exact location of damage has been problematic for researchers. Nevertheless, detecting the failure of lateral supports in various parts of a structure using time–frequency analysis and deep learning methods, such as convolutional neural networks, is possible. Then, the damaged structure can be rapidly rebuilt to ensure safety. Experiments are conducted to determine the vibration acceleration modes of a four-storey steel structure considering various support structure damage scenarios. The acceleration signals at each measurement point are then analysed with respect to time and frequency to generate appropriate three-dimensional spectral matrices. In this study, the MobileNetV2 deep learning model was trained on a labelled picture collection of damaged matrix images. Hyperparameter tweaking and training resulted in a prediction accuracy of 97.37% for the complete dataset and 99.30% and 96.23% for the training and testing sets, respectively. The findings indicate that a combination of time–frequency analysis and deep learning methods may pinpoint the position of the damaged steel frame support components more accurately.

show abstract

Section: Introductionmentioning

confidence: 99%

Deep Learning Damage Identification Method for Steel- Frame Bracing Structures Using Time–frequency Analysis and Convolutional Neural Networks

2023

View full text Add to dashboard Cite

show abstract

“…This idea of easier detection by using a variety of moving loads in the analysis was inspired by the results of a study by Chouiyakha et al [17] as well as the results of a study by Roveri and Carcaterra [18] that demonstrated that their proposed methods based on moving load were truly able to identify the locations of the cracks precisely. The approaches based on moving load for damage detection in structures were widely used by several researchers in the past [19][20][21][22]. In contrast, our proposed method would be based on vibrational data in time domain instead of vibrational data in frequency domain (used in most of the studies mentioned above) to counter a common problem for many engineers that is responsible for crack identification of their facilities: complex and time-consuming tasks encoding unnecessarily complex mathematical calculation steps into a functioning and precise crack-identification app for their intended structure.…”

Section: Introductionmentioning

confidence: 99%

Crack Identification in Cantilever Beam under Moving Load Using Change in Curvature Shapes

et al. 2022

View full text Add to dashboard Cite

Cracks in structural components may ultimately lead to failure of the structure if not identified sufficiently early. This paper presents a crack-identification method based on time-domain. Captured time-domain data are processed into central difference approximation of displacement of each node (point) in the structure. Abnormally high central difference approximation of displacement of a node relative to those of its neighborhood points indicates a crack at that point. A suite of simulation experiments and numerical calculations was conducted to find out whether the proposed identification method could accurately identify the location of a crack in a cantilever beam under moving load compared to the location found by an exact solution method, and the outcomes indicated that it was as able as the analytical method. The proposed method is an FEA analysis, an approach familiar to virtually every engineer. Therefore, the relative amount of time and effort spent on developing the proposed method for a specific application is much less than those spent on developing an analytical method. The saved time and effort should enable more engineering personnel to perform routine checks on structural elements of their interest more simply and frequently.

show abstract

“…[53] Some different versions of ELM have been applied for structural damage identification. [54][55][56] This present study applies newly developed optimization algorithms of multiverse optimizer (MVO), sine cosine algorithm (SCA), and Harris hawks optimization (HHO) for structural damage identification. [57][58][59] MVO has already been used for optimal tuning of fuzzy parameters for structural motion control.…”

Section: Introductionmentioning

confidence: 99%

“…[ 53 ] Some different versions of ELM have been applied for structural damage identification. [ 54–56 ]…”

Section: Introductionmentioning

confidence: 99%

Multiverse optimizer for structural damage detection: Numerical study and experimental validation

Ghannadi

Kourehli

2020

Structural Design Tall Build

Self Cite

View full text Add to dashboard Cite

New optimization algorithms have been increasingly developed in the course of the recent years. The accuracy of identified results in optimization-based damage detection methods depends on the objective function and optimization algorithm. This paper employs multiverse Optimizer (MVO) to solve the optimization-based damage identification problem. Statistical results obtained by MVO are compared with those of sine cosine algorithm (SCA) and Harris hawks optimization (HHO) in order to perform a comparative study. Two objective functions are used in this optimization problem. The first one is based on the modal assurance criterion (MAC) and the second one on modified total modal assurance criterion (MTMAC). Numerical and experimental examples indicate that the combination of objective function based on MTMAC and MVO algorithm can provide accurate and reliable results in structural damage detection process. MAC is also rejected in competition with MTMAC. While using SCA-MTMAC, the accuracy of the results is acceptable in some cases. However, HHO-MTMAC gives weak results in most cases. MVO and SSA are the lowest, and HHO spends significantly more time in terms of computational cost.

show abstract

Vibration analysis and identification of breathing cracks in beams subjected to single or multiple moving mass using online sequential extreme learning machine

Cited by 9 publications

References 18 publications

Deep Learning Damage Identification Method for Steel- Frame Bracing Structures Using Time–frequency Analysis and Convolutional Neural Networks

Deep Learning Damage Identification Method for Steel- Frame Bracing Structures Using Time–frequency Analysis and Convolutional Neural Networks

Crack Identification in Cantilever Beam under Moving Load Using Change in Curvature Shapes

Multiverse optimizer for structural damage detection: Numerical study and experimental validation

Contact Info

Product

Resources

About