Acoustic emission for in situ quality monitoring in additive manufacturing using spectral convolutional neural networks

Shevchik, Sergey; Kenel, Christoph; Leinenbach, Christian; Wasmer, Kilian

doi:10.1016/j.addma.2017.11.012

Cited by 226 publications

(135 citation statements)

References 21 publications

Supporting

Mentioning

132

Contrasting

Unclassified

Order By: Relevance

“…Previous studies have shown that machine learning can be deployed for defect detection in the L-PBF process, using various techniques like acoustic sensors [34], thermal [36], grayscale [32], or high-resolution imaging [31]. While the current benchmarks range from 77 to 98% accuracy in detecting errors, they rely on either extensive data preprocessing or upon additional imaging techniques.…”

Section: Discussionmentioning

confidence: 99%

See 1 more Smart Citation

A deep learning-based model for defect detection in laser-powder bed fusion using in-situ thermographic monitoring

et al. 2020

View full text Add to dashboard Cite

Additive manufacturing of metal components with laser-powder bed fusion is a very complex process, since powder has to be melted and cooled in each layer to produce a part. Many parameters influence the printing process; however, defects resulting from suboptimal parameter settings are usually detected after the process. To detect these defects during the printing, different process monitoring techniques such as melt pool monitoring or off-axis infrared monitoring have been proposed. In this work, we used a combination of thermographic off-axis imaging as data source and deep learning-based neural network architectures, to detect printing defects. For the network training, a k-fold cross validation and a hold-out cross validation were used. With these techniques, defects such as delamination and splatter can be recognized with an accuracy of 96.80%. In addition, the model was evaluated with computing class activation heatmaps. The architecture is very small and has low computing costs, which means that it is suitable to operate in real time even on less powerful hardware.

show abstract

Section: Discussionmentioning

confidence: 99%

“…The accuracy of the algorithm was at 77% and reduced experimental data are needed for training compared to supervised ML. Shevchik et al [34] and Ye et al [35] used acoustic signals for defect detection with ML. The acoustic signals need more data preparation before they can be used in algorithm.…”

Section: Related Workmentioning

confidence: 99%

A deep learning-based model for defect detection in laser-powder bed fusion using in-situ thermographic monitoring

et al. 2020

View full text Add to dashboard Cite

show abstract

“…which, together with different fault categories, serve as input to train and test a deep CNN in terms of sideband amplitudes with an accuracy of 99.5%. Another successful approach that combines acoustic emission sensor signals converted to time-frequency spectrograms for real-time monitoring of the workpiece quality with CNNs was reported in [28]. The application of a hybrid feature extraction method combined with CNNs for tool condition monitoring was reported in [29], where signals from the dynamometer and accelerometer were independently subject to time-frequency transformation using Morlet wavelet transform and feature extraction using PyWavelets.…”

Section: Deep Learning In Condition Monitoringmentioning

confidence: 99%

Online Tool Wear Classification during Dry Machining Using Real Time Cutting Force Measurements and a CNN Approach

Terrazas

Martínez-Arellano

Benardos

et al. 2018

JMMP

View full text Add to dashboard Cite

The new generation of ICT solutions applied to the monitoring, adaptation, simulation and optimisation of factories are key enabling technologies for a new level of manufacturing capability and adaptability in the context of Industry 4.0. Given the advances in sensor technologies, factories, as well as machine tools can now be sensorised, and the vast amount of data generated can be exploited by intelligent information processing techniques such as machine learning. This paper presents an online tool wear classification system built in terms of a monitoring infrastructure, dedicated to perform dry milling on steel while capturing force signals, and a computing architecture, assembled for the assessment of the flank wear based on deep learning. In particular, this approach demonstrates that a big data analytics method for classification applied to large volumes of continuously-acquired force signals generated at high speed during milling responds sufficiently well when used as an indicator of the different stages of tool wear. This research presents the design, development and deployment of the system components and an overall evaluation that involves machining experiments, data collection, training and validation, which, as a whole, has shown an accuracy of 78%.

show abstract

“…The usage of sensors, allied with the digital signal processing to monitor the additive manufacturing processes, can be found in a large number of applications. For instance, acoustic emission (AE) sensors are applied in the monitoring of Selective Laser Melting (SLM) [3], Laser Metal Deposition (LMD) [4] and FDM manufacturing processes [2,5], as well as accelerometers are used for monitoring of FDM process [6].…”

Section: Introductionmentioning

confidence: 99%

Evaluating Temperature Influence on Low-Cost Piezoelectric Transducer Response for 3D Printing Process Monitoring

Lopes

Rocha

Aguiar

et al. 2019

The 6th International Electronic Conference on Sensors and Applications

View full text Add to dashboard Cite

The 3D printing process deals with the production of three-dimensional objects with defined geometries. However, this manufacturing process has a crucial point established at the beginning of the object manufacturing, where anomalies can occur and compromise the entire object produced. The piezoelectric diaphragm has been studied as an alternative to the conventional acoustic emission (AE) sensor concerning the monitoring of structures and processes. It has in its assembling a ceramic element with piezoelectric properties, which makes its response sensitive to temperature variations. The Pencil Lead Break (PLB) method is widely used due to its efficiency in the characterization of AE sensors. The present work aims to study the influence of temperature on the piezoelectric diaphragm response for the monitoring of the 3D printing process. PLB tests were performed on the glass surface of a 3D printer at three different temperatures, and the raw signal was collected at 5 MHz sample rate. The signal was investigated in the time and frequency domain. The results demonstrate that the frequency response of the sensor is directly influenced by the temperature variations. In addition, the signal amplitude variations occur differently along the entire spectrum, and frequency bands with small and large amplitude variations can be selected for a comparison study. Furthermore, two frequency bands were carefully selected, and the mean error was obtained regarding the reference temperatures of 25 °C and 45 °C. It can be inferred that the piezoelectric transducer has low sensitivity to temperature variation if a proper frequency band is selected, where an acceptable error of 16.9% was obtained.

show abstract

Acoustic emission for in situ quality monitoring in additive manufacturing using spectral convolutional neural networks

Cited by 226 publications

References 21 publications

A deep learning-based model for defect detection in laser-powder bed fusion using in-situ thermographic monitoring

A deep learning-based model for defect detection in laser-powder bed fusion using in-situ thermographic monitoring

Online Tool Wear Classification during Dry Machining Using Real Time Cutting Force Measurements and a CNN Approach

Evaluating Temperature Influence on Low-Cost Piezoelectric Transducer Response for 3D Printing Process Monitoring

Contact Info

Product

Resources

About