On the application of machine learning for defect detection in L-PBF additive manufacturing

Mohammadi, Mohammad; Mahmoud, Dalia; Elbestawi, M.A.

doi:10.1016/j.optlastec.2021.107338

Cited by 57 publications

(9 citation statements)

References 25 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Machine learning also plays a vital role in the defect detection of additive manufacturing. Ghayoomi Mohammadi et al [ 24 ] applied various machine learning algorithms to achieve real-time defect detection during laser powder additive printing. The data processed by the machine learning algorithm and the training data in the dataset were both generated by the continuous monitoring of the printing by the acoustic emission sensor.…”

Section: Additive Manufacturing With Machine Learningmentioning

confidence: 99%

Machine learning boosts three-dimensional bioprinting

Ning¹,

Zhou²,

Joo³

2023

IJB

View full text Add to dashboard Cite

Three-dimensional (3D) bioprinting is a computer-controlled technology that combines biological factors and bioinks to print an accurate 3D structure in a layer-by-layer fashion. 3D bioprinting is a new tissue engineering technology based on rapid prototyping and additive manufacturing technology, combined with various disciplines. In addition to the problems in in vitro culture process, the bioprinting procedure is also afflicted with a few issues: (1) difficulty in looking for the appropriate bioink to match the printing parameters to reduce cell damage and mortality; and (2) difficulty in improving the printing accuracy in the printing process. Data-driven machine learning algorithms with powerful predictive capabilities have natural advantages in behavior prediction and new model exploration. Combining machine learning algorithms with 3D bioprinting helps to find more efficient bioinks, determine printing parameters, and detect defects in the printing process. This paper introduces several machine learning algorithms in detail, summarizes the role of machine learning in additive manufacturing applications, and reviews the research progress of the combination of 3D bioprinting and machine learning in recent years, especially the improvement of bioink generation, the optimization of printing parameter, and the detection of printing defect.

show abstract

Section: Additive Manufacturing With Machine Learningmentioning

confidence: 99%

Machine learning boosts three-dimensional bioprinting

Ning¹,

Zhou²,

Joo³

2023

IJB

View full text Add to dashboard Cite

show abstract

“…Li et al [10] proposed measuring part quality using image data collected during the process to build a deep learning-based quality identification method. Mohammadi et al [11] studied reliable product quality through dimensional error prediction, investigating the performance of several different ML methodologies to detect defects in real-time. Aminzadeh et al [12] researched the inspection of a parts' dimensional accuracy during the build process through the development of machine-vision-based, dimensionalinspection techniques.…”

Section: Background and Literature Review Of Modelingmentioning

confidence: 99%

Ensuring Quality in Metal Additive Manufacturing Through a V-Model Framework

Roh,

Simpson,

Yang

et al. 2023

IEEE Access

View full text Add to dashboard Cite

Metal Additive Manufacturing (MAM) produces complex, part geometries from a variety of materials in powder and wire form. Due to complexities of MAM processes that create those geometries, especially powder bed fusion, quality assurance, and qualification remain an ongoing challenge. Quality assurance involves assessing the quality of a part's geometry, surface finish, and mechanical properties. Currently, quality assurance is not easily achieved due to variations in the powder inputs, the MAM process itself, and environmental factors, such as temperature. Many efforts are underway to develop a new quality system that includes 1) planning, measuring, and qualifying parts and 2) enhancing quality through a processing-monitoring-qualifying framework. Creating this new system requires building the complex relationships between requirements, processes, and quality. These relationships are needed to specify, measure, analyze, and optimize variables to ensure final part quality. Thus, a processing-monitoring-quality framework could provide critical steps to identify those relationships and help meet stakeholder needs. The paper describes how to adapt the "software and systems engineering" V-model to a "metal AM quality assurance" V-model that can provide a framework for quality assurance in MAM.

show abstract

“…Deep learning techniques have been employed in anomaly detection in various fields 6 – 11 We developed an anomaly detection module that identifies signal bond defects from images captured using the railway monitoring system installed on commercial trains of the East Japan Railway Company (JR-EAST). Because the images captured by the railway monitoring system involve multiwires and odds, which could interfere with anomaly detection, the EfficientDet 12 -based object detection model was employed for extracting signal bond parts.…”

Section: Introductionmentioning

confidence: 99%

Employing a hybrid model based on texture-biased convolutional neural networks and edge-biased vision transformers for anomaly detection of signal bonds

2023

View full text Add to dashboard Cite

The railway system of Japan plays a vital role in the national transportation network.A key issue in public transport safety is anomaly detection in railways. Lately, developing robust algorithms and methods for anomaly detection has become the premier task in this field. Recently introduced approaches based on convolutional neural networks, generative adversarial networks, and vision transformers (ViTs) have remarkably improved the research in anomaly detection. Our work proposes a high-performance module for the anomaly detection of signal bonds. First, we present an overview of the proposed module; then, the object detection model and the proposed hybrid classification model based on texture-biased convolutional neural networks and edge-biased ViTs are introduced. Finally, the proposed anomaly detection module is evaluated for accuracy using the dataset from the East Japan Railway Company and the receiver operating characteristic curve. The results show that our proposed module achieves a better performance on real-life data than the two methods that were combined to generate it, raising hope toward possible usage in other areas as well.

show abstract

On the application of machine learning for defect detection in L-PBF additive manufacturing

Cited by 57 publications

References 25 publications

Machine learning boosts three-dimensional bioprinting

Machine learning boosts three-dimensional bioprinting

Ensuring Quality in Metal Additive Manufacturing Through a V-Model Framework

Employing a hybrid model based on texture-biased convolutional neural networks and edge-biased vision transformers for anomaly detection of signal bonds

Contact Info

Product

Resources

About