DeepSHM: a deep learning approach for structural health monitoring based on guided Lamb wave technique

Ewald, Vincentius; Groves, Roger M.; Benedictus, Rinze

doi:10.1117/12.2506794

Cited by 48 publications

(40 citation statements)

References 30 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…By using decision tree methods or random cross-validation with different features combinations, one could rank the most important features, and use these to inform the approach taken, saving time and providing simpler solutions. From the ML viewpoint, a data augmentation approach [41], [66] can be addressed for both the simulated and experimental datasets of this work. Another approach would be addressing each of the mode separation datasets…”

Section: Discussionmentioning

confidence: 99%

“…Melville et al [40] employed convolutional neural networks to estimate the damage level of a thin metal plate, emulated by a steel washer, using full wavefield measured data. DL is also explored with Lamb waves, in which, wavelet coefficients are extracted and used as inputs of the models [41]. Moreover, Hesser et al [42] have proposed an active source localization to predict the impact position of a steel ball in an aluminium plate, using numerical simulations to train and experimental data to validate ML models.…”

Section: Introductionmentioning

confidence: 99%

“…Content may change prior to final publication. with automatic feature extraction -in other words DL techniques [39] that could use for example convolutional neural networks [40], [41].…”

mentioning

confidence: 99%

See 2 more Smart Citations

Machine Learning-Based Corrosion-Like Defect Estimation With Shear-Horizontal Guided Waves Improved by Mode Separation

et al. 2021

View full text Add to dashboard Cite

Shear Horizontal (SH) guided waves have been extensively used to estimate and detect defects in structures like plates and pipes. Depending on the frequency and plate thickness, more than one guided-wave mode propagates, which renders signal interpretation complicated due to mode mixing and complex behavior of each individual mode interacting with defects. This paper investigates the use of machine learning models to analyse the two lowest order SH guided modes, for quantitative size estimation and detection of corrosion-like defects in aluminium plates. The main contribution of the present work is to show that mode separation through machine learning improves the effectiveness of predictive models. Numerical simulations have been performed to generate time series for creating the estimators, while experimental data have been used to validate them. We show that a full mode separation scheme decreased the error rate of the final model by 30% and 67% in defect size estimation and detection respectively.

show abstract

Section: Discussionmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation

Machine Learning-Based Corrosion-Like Defect Estimation With Shear-Horizontal Guided Waves Improved by Mode Separation

et al. 2021

View full text Add to dashboard Cite

show abstract

“…In addition to CS, other techniques in data science and engineering also attract attention in the field of structural health monitoring [31]. A novel framework called DeepSHM was presented to classify damage using deep learning (DL) [32]. It trained the neural weights by wavelet coefficient matrix and extracted feature patterns.…”

Section: Related Workmentioning

confidence: 99%

Compressed Sensing Method for Health Monitoring of Pipelines Based on Guided Wave Inspection

Wang

Huang

Wang

et al. 2020

IEEE Trans. Instrum. Meas.

View full text Add to dashboard Cite

The pipeline in-service needs to be inspected in certain period to master its structural health status. Ultrasonic guided wave which can propagate along pipelines with less energy loss provides an efficient method for long-term in situ inspection. Guided wave can detect both corrosion and cracks existing in structures. To overcome the problem of huge amounts of data and to maintain defect identification accuracy, the compressed sensing method for guided wave inspection is proposed. The compression process is essentially a scheme of analog to information conversion to compress the signal. It is accomplished by random demodulation and the equivalent sampling rate below the Nyquist rate helps to save most of storage. The compressed data is recovered to the sparse spatial domain based on the constructed dictionary from a guided wave propagation model. To verify the effectiveness of proposed method, both numerical simulations and experimental investigations are conducted. The results indicate the availability of compression and high accuracy of defect location after recovery. The influences of different compression schemes and compression ratios are further analyzed. In addition, the comparisons with direct recovery without compression and traditional analysis methods demonstrate the advantageous performance of the proposed method.

show abstract

“…existence, location, size, and type of damage) which is dependent on the captured signal X Π , T , Ψ , Ω , and influenced by parameter tuples Π , T , Ψ , Ω , which correspond to actor, transitional, medium, and environmental domain, respectively. A detailed explanation of these parameters can be found in Ewald et al, 25 and to summarize we depict the above-mentioned parameters in Figure 1. Equation (1) can be understood by considering that the posterior belief P ( h θ ( X Π , T , Ψ , Ω )| X Π , T , Ψ , Ω ) is equal to the multiplication of the prior belief P ( X Π , T , Ψ , Ω ) by the likelihood P ( X Π , T , Ψ , Ω | h θ ( X Π , T , Ψ , Ω) ) that X Π , T , Ψ , Ω will occur given that h θ is true.…”

Section: Theorymentioning

confidence: 99%

Integrative approach for transducer positioning optimization for ultrasonic structural health monitoring for the detection of deterministic and probabilistic damage location

Ewald

Groves

Benedictus

2020

Structural Health Monitoring

Self Cite

View full text Add to dashboard Cite

The concept of structural health monitoring has been introduced to ensure structural integrity during the design lifetime of a structure. The main objectives of structural health monitoring are to detect, locate, quantify, and predict any damage that occurs during this lifetime of the structure so that effective and efficient maintenance and repair procedures can be performed. The location of structural damage events can be discretized as deterministic and probabilistic. A deterministic location specifies that the damage occurs in high-stress regions or other regions that can be predicted by the structural design, such as the most probable location for a fatigue crack. A probabilistic damage event is one where the location of the damage is independent of structural design parameters, such as hail impact, bird strike, and impact from ground vehicles. A structural health monitoring system should be able to handle both these damage occurrences. In our previous work, we optimized the transducer placement in Lamb wave–based structural health monitoring for the detection of a fatigue crack that emerges from a rivet hole. In this article, we demonstrate a combination of that method with a different sensor placement optimization method to add the capability to detect probabilistic damage location. First, we considered the ultrasonic wave attenuation in the structure and based on this attenuation, we created a fitness function. Since this fitness function is difficult to solve due to its combinatorial nature, we compared three common metaheuristic stochastic strategies: global random search, greedy algorithm, and genetic algorithm, for solving this problem. The results of this analysis were then integrated with the previously described deterministic approach, making a global structural health monitoring sensor placement strategy that balances the need to detect both pre-determined and random damage location occurrences. The analytical result of the study presented is validated by experiment.

show abstract

DeepSHM: a deep learning approach for structural health monitoring based on guided Lamb wave technique

Cited by 48 publications

References 30 publications

Machine Learning-Based Corrosion-Like Defect Estimation With Shear-Horizontal Guided Waves Improved by Mode Separation

Machine Learning-Based Corrosion-Like Defect Estimation With Shear-Horizontal Guided Waves Improved by Mode Separation

Compressed Sensing Method for Health Monitoring of Pipelines Based on Guided Wave Inspection

Integrative approach for transducer positioning optimization for ultrasonic structural health monitoring for the detection of deterministic and probabilistic damage location

Contact Info

Product

Resources

About