Powder Bed Fusion via Machine Learning-Enabled Approaches

Complexity, Journal Year: 2023, Volume and Issue: 2023, P. 1 - 25

Published: April 30, 2023

Powder bed fusion (PBF) applies to various metallic materials used in the metal printing process of building a wide range complex parts compared other AM technologies. PBF has several variants such as DMLS (direct laser sintering), EBM (electron beam melting), SHS (selective heat SLM and SLS sintering). For reach its maximum potential, machine learning (ML) algorithms are with suitable achieve goals cost-effectively. Various applications neural networks, including ANNs, CNNs, RNNs, popular techniques KNN, SVM, GP were reviewed, future challenges discussed. Some special-purpose listed follows: GAN, SeDANN, SCNN, K-means, PCA, etc. This review presents evolution, current status, challenges, prospects these technologies terms material, features, parameters, applications, advantages, disadvantages, etc., explain their significance provide an in-depth understanding same.

Language: Английский

---

Pulsed excitation heating of semiconductor material and its thermomagnetic response on the basis of fourth-order MGT photothermal model

Acta Mechanica, Journal Year: 2023, Volume and Issue: 234(10), P. 4977 - 4995

Published: July 10, 2023

Language: Английский

---

Deep learning-based framework for the observation of real-time melt pool and detection of anomaly in wire-arc additive manufacturing

Materials and Manufacturing Processes, Journal Year: 2023, Volume and Issue: 39(6), P. 761 - 777

Published: Sept. 5, 2023

ABSTRACTObject detection has become a popular tool of deep learning in the era digital manufacturing. In this study, most powerful and efficient object algorithm, i.e., You Only Look Once (YOLO) was used to detect anomalies deposited beads wire-arc additive manufacturing (WAAM) using melt pool images. This study latest version YOLO algorithm train validate custom image dataset obtained by conducting experiments robotic-controlled WAAM. The mean average precision (mAP) for "Regular bead" class "Irregular reached 99% at an Intersection over Union (IoU) threshold 0.5, both training validation. When model tested new or unseen datasets four experimental trials, mAP value 98.47% 96.68% processing time 0.014 s/frame. shown excellent 15 ms per frame, which shows its potential real-time application industry.KEYWORDS: WAAMdeep learningobject detectionYOLOv8real-time AcknowledgmentsThe authors would like thank Department Production Industrial Engineering, BIT Sindri, Dhanbad providing research facility.Disclosure statementNo conflict interest reported author(s).

Language: Английский

---

Advancing Additive Manufacturing through Deep Learning: A Comprehensive Review of Current Progress and Future Challenges

IISE Transactions, Journal Year: 2024, Volume and Issue: unknown, P. 1 - 44

Published: Dec. 19, 2024

This paper presents the first comprehensive literature review of deep learning (DL) applications in additive manufacturing (AM). It addresses need for a thorough analysis this rapidly growing yet scattered field, aiming to bring together existing knowledge and encourage further development. Our research questions cover three major areas AM: (i) design AM, (ii) AM modeling, (iii) monitoring control AM. We use step-by-step approach following Preferred Reporting Items Systematic Reviews Meta-Analyses (PRISMA) guidelines select papers from Scopus Web Science databases, aligning with our questions. include only those that implement DL across seven categories - binder jetting, directed energy deposition, material extrusion, powder bed fusion, sheet lamination, vat photopolymerization. reveals trend towards using generative models, such as adversarial networks, also highlights an increasing effort incorporate process physics into models improve modeling reduce data requirements. Additionally, there is interest 3D point cloud monitoring, alongside traditional 1D 2D formats. Finally, summarizes current challenges recommends some promising opportunities domain investigation special focus on generalizing wide range geometry types, managing uncertainties both overcoming limited, imbalanced, noisy issues by incorporating (iv) unveiling potential interpretable

Language: Английский

---

Thermal Frequency Analysis of Double CNT-Reinforced Polymeric Straight Beam

Journal of Vibration Engineering & Technologies, Journal Year: 2023, Volume and Issue: 12(1), P. 649 - 665

Published: Jan. 26, 2023

Language: Английский

---

Barriers to Adoption of Artificial Intelligence in Metal Additive Manufacturing

Published: Jan. 14, 2025

<div class="section abstract"><div class="htmlview paragraph">Artificial intelligence (AI) is poised to significantly impact metal additive manufacturing (AM). Understanding how one might use AI in AM challenging because experts are not experts, nor the other way around. This document introduces and guides researchers accessing relevant literature. It also discusses hype surrounding AM, rush publish peer-reviewed papers that resulting uneven quality of Conclusions regarding application both large small enterprises discussed.</div><div paragraph">This intended help illuminate for<ul class="list disc"><li class="list-item"><div paragraph">Hands-on engineers who need quickly understand what levels problems they encounter when dealing with AM</div></li><li paragraph">Engineering managers stay current on emerging trends their technical realm responsibilities</div></li><li paragraph">Policymakers may have expertise</div></li><li paragraph">Faculty students want an introduction AM</div></li></ul></div><div paragraph">NOTE: SAE Edge Research Reports identify key issues emerging, but still unsettled, technologies interest mobility industry. The goal stimulate discussion work hope promoting speeding resolution identified issues. These reports resolve challenges or close any topic further scrutiny.</div></div>

Language: Английский

---

Learning with limited annotations: Deep semi-supervised learning paradigm for layer-wise defect detection in laser powder bed fusion

Optics & Laser Technology, Journal Year: 2025, Volume and Issue: 185, P. 112586 - 112586

Published: Feb. 17, 2025

Language: Английский

---

In-Situ monitoring of porosity based on static and dynamic molten pool features in laser powder bed fusion

Optics & Laser Technology, Journal Year: 2025, Volume and Issue: 187, P. 112872 - 112872

Published: March 30, 2025

Language: Английский

---

Femtosecond laser-made 3D micro-chainmail scaffolds towards regenerative medicine

Optics & Laser Technology, Journal Year: 2023, Volume and Issue: 162, P. 109240 - 109240

Published: Feb. 15, 2023

Language: Английский

---

Additively Manufactured Carbon-Reinforced ABS Honeycomb Composite Structures and Property Prediction by Machine Learning

Molecules, Journal Year: 2024, Volume and Issue: 29(12), P. 2736 - 2736

Published: June 8, 2024

The expansive utility of polymeric 3D-printing technologies and demand for high- performance lightweight structures has prompted the emergence various carbon-reinforced polymer composite filaments. However, detailed characterization processing-microstructure-property relationships these materials is still required to realize their full potential. In this study, acrylonitrile butadiene styrene (ABS) two ABS variants, with either carbon nanotubes (CNT) or 5 wt.% chopped fiber (CF), were designed in a bio-inspired honeycomb geometry. These manufactured by fused filament fabrication (FFF) investigated across range layer thicknesses hexagonal (hex) sizes. Microscopy material cross-sections was conducted evaluate relationship between print parameters porosity. Analyses determined trend reduced porosity lower print-layer heights hex sizes compared larger Mechanical properties evaluated through compression testing, specimens achieving higher compressive yield strength, while CNT-ABS achieved ultimate strength due reduction subsequent strengthening. A decreasing increasing size all supported negative correlation height size. We elucidated potential ABS, CNT-ABS, ABS-5wt.% CF composites novel 3D-printed structures. studies development predictive classification regression supervised machine learning model 0.92 accuracy 0.96 coefficient determination help inform guide design targeted performance.

Language: Английский

---