CNN-LSTM network-based damage detection approach for copper pipeline using laser ultrasonic scanning

Huang, Liuwei; Hong, Xiaobin; Yang, Zhijing; Liu, Yuan; Zhang, Bin

doi:10.1016/j.ultras.2022.106685

Cited by 56 publications

(9 citation statements)

References 41 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…CNNs are primarily used for the feature recognition of two-dimensional images, whereas 1D-CNNs have only one dimension; therefore, they are widely used for feature recognition and time-series extraction. Although a 1D-CNN only has a single dimension, it demonstrates the same advantages as a CNN [ 14 ]. Specifically, CNNs not only present the advantages of traditional neural networks, such as a strong self-learning ability and good adaptability, but also the advantages of weight sharing and easy model training [ 23 ].…”

Section: Defect Quantification Methods Based On Ultrasonic Guided Wav...mentioning

confidence: 99%

“…A CNN network that extracts information directly from the ultrasound-guided wave response signal is highly effective in avoiding this issue. Huang [ 14 ] proposed a damage detection method based on a CNN-LSTM network for laser ultrasonic guided wave scanning detection. Miorelli et al [ 15 ] proposed an automatic method for localizing and quantifying structural health monitoring defects based on guided wave imaging by combining convolutional neural networks.…”

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation

Quantitative Detection of Pipeline Cracks Based on Ultrasonic Guided Waves and Convolutional Neural Network

Shen,

Wu,

Chen

et al. 2024

Sensors

View full text Add to dashboard Cite

In this study, a quantitative detection method of pipeline cracks based on a one-dimensional convolutional neural network (1D-CNN) was developed using the time-domain signal of ultrasonic guided waves and the crack size of the pipeline as the input and output, respectively. Pipeline ultrasonic guided wave detection signals under different crack defect conditions were obtained via numerical simulations and experiments, and these signals were input as features into a multi-layer perceptron and one-dimensional convolutional neural network (1D-CNN) for training. The results revealed that the 1D-CNN performed better in the quantitative analysis of pipeline crack defects, with an error of less than 2% in the simulated and experimental data, and it could effectively evaluate the size of crack defects from the echo signals under different frequency excitations. Thus, by combining the ultrasonic guided wave detection technology and CNN, a quantitative analysis of pipeline crack defects can be effectively realized.

show abstract

Section: Defect Quantification Methods Based On Ultrasonic Guided Wav...mentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

Quantitative Detection of Pipeline Cracks Based on Ultrasonic Guided Waves and Convolutional Neural Network

Shen,

Wu,

Chen

et al. 2024

Sensors

View full text Add to dashboard Cite

show abstract

“…Signal extraction Deep learning methods have gained significant attention with the increasing volume of data and the growing complexity of ultrasonic signals. Common deep learning networks include convolutional neural networks (CNN) [104], graph neural networks (GNNs) [98], recurrent neural networks (RNN) [105], long short-term memory (LSTM) [106], and Au-toEncoder [107]. It is evident that data-driven techniques dominated by machine learning (ML) methods have demonstrated significant advantages in ultrasonic in-line inspection compared to physical models [108].…”

Section: Potential Challenges and Opportunitiesmentioning

confidence: 99%

Systematic Evaluation of Ultrasonic In-Line Inspection Techniques for Oil and Gas Pipeline Defects Based on Bibliometric Analysis

Huang,

Chen,

et al. 2024

Sensors

View full text Add to dashboard Cite

The global reliance on oil and gas pipelines for energy transportation is increasing. As the pioneering review in the field of ultrasonic defect detection for oil and gas pipelines based on bibliometric methods, this study employs visual analysis to identify the most influential countries, academic institutions, and journals in this domain. Through cluster analysis, it determines the primary trends, research hotspots, and future directions in this critical field. Starting from the current global industrial ultrasonic in-line inspection (ILI) detection level, this paper provides a flowchart for selecting detection methods and a table for defect comparison, detailing the comparative performance limits of different detection devices. It offers a comprehensive perspective on the latest ultrasonic pipeline detection technology from laboratory experiments to industrial practice.

show abstract

“…Melville et al used deep learning (DL) to achieve accurate detection of damage using only partial (0.1%) wavefield measurement [268]. Huang et al proposed a CNN-LSTM (convolutional neural network-long short-term memory) networkbased damage detection approach [269]. The CNN-LSTM network not only extracts the signal features but also retains the signal time-domain characteristics in order to improve detection accuracy.…”

Section: • Resonant Frequency Identificationmentioning

confidence: 99%

Guided ultrasonic waves propagation imaging: a review

Chia¹,

Lee²,

Harmin³

et al. 2023

Meas. Sci. Technol.

View full text Add to dashboard Cite

This article presents a comprehensive review of the laser-based guided ultrasonic waves propagation imaging (G-UPI) system and respective signal/data processing methods related to the nondestructive testing and evaluation of thin-walled structures. The primary goal of this study is to review and recognize various processing methods, explain the working principles of the most influential methods, and highlight outstanding capabilities. In addition, the suitability of the methods for multiple types of damage and defect in various materials and structures are presented. At the same time, success stories of difficult-to-inspect cases are highlighted. Its secondary goal is to compare and discuss the merits and demerits of the laser-scanning part of the system for ultrasound generation and acquisition to provide a guideline for scanning scheme or hardware selection. Finally, the potential challenges and prospects of the G-UPI are discussed. It is expected that this review would serve as an entrance key for newcomers and a reference point for researchers to explore the opportunities for further improvement in the laser ultrasound-based evaluation of critical engineering structures.

show abstract

CNN-LSTM network-based damage detection approach for copper pipeline using laser ultrasonic scanning

Cited by 56 publications

References 41 publications

Quantitative Detection of Pipeline Cracks Based on Ultrasonic Guided Waves and Convolutional Neural Network

Quantitative Detection of Pipeline Cracks Based on Ultrasonic Guided Waves and Convolutional Neural Network

Systematic Evaluation of Ultrasonic In-Line Inspection Techniques for Oil and Gas Pipeline Defects Based on Bibliometric Analysis

Guided ultrasonic waves propagation imaging: a review

Contact Info

Product

Resources

About