Research and application of artificial intelligence techniques for wire arc additive manufacturing: a state-of-the-art review

He, Fengyang; Yuan, Lei; Mu, Haochen; Ros, Montserrat; Ding, Donghong; Pan, Zengxi; Li, Huijun

doi:10.1016/j.rcim.2023.102525

Cited by 35 publications

(16 citation statements)

References 110 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Ko et al [8] constructed physics-based models using graph-based networks for laser powder bed fusion processes, indicating the possibility of a knowledge-based approach to investigate any potential accuracy improvements. He et al [9] also corroborate the popularity of neural networks and support vector machines among the research works published in the field of wire arc additive manufacturing. They noted that support vector machines can be trained on small datasets to obtain reasonably good prediction accuracies, giving a solution for dataset availability problems.…”

Section: Introductionmentioning

confidence: 64%

Anomaly Detection in WAAM Deposition of Nickel Alloys—Single-Material and Cross-Material Analysis

Rajesh,

Ya,

Hermans

2023

Metals

View full text Add to dashboard Cite

The current research work investigates the possibility of using machine learning models to deduce the relationship between WAAM (wire arc additive manufacturing) sensor responses and defect presence in the printed part. The work specifically focuses on three materials from the nickel alloy family (Inconel 718, Invar 36 and Inconel 625) and uses three sensor responses for data analysis, which are welding voltage, welding current and welding audio. Two different machine learning models are used—artificial neural networks (ANNs) and random forests (RF). The results for each of the materials, separately, indicate that the accuracies range from 60% to 90% and the correlation coefficient is less than 0.5 (indicating weak positive correlation), depending on the model and material. In addition to separate material analysis, a cross-material data analysis was formed to test the models’ general prediction capabilities. This led to predictions that are significantly worse, with accuracies ranging from 20% to 27% and very weak correlation coefficients (less than 0.1), indicating that the choice of material is still important as a boundary condition. Analysis of the results indicates that the relative importance of audio sensor response depends on the nature of defect formation. Random forests are found to perform the best for single material analysis, with the comparatively inferior performance of ANNs possibly being due to lack of sufficient datapoints.

show abstract

Section: Introductionmentioning

confidence: 64%

Anomaly Detection in WAAM Deposition of Nickel Alloys—Single-Material and Cross-Material Analysis

Rajesh,

Ya,

Hermans

2023

Metals

View full text Add to dashboard Cite

show abstract

“…The temperature of the molten pool in L-DED process play an important role, as reported by Tang et al (2020) as it affects the forming appearance of the deposited layer, with the phenomenon explained by forming failure and edge defects that occurred in a thin-wall accumulation of a cylinder cladding. In recent work (He et al, 2023) reported several techniques, using physical sensors and melt pool image data to be used as input for online control of the deposition process, since the molten pool dynamics directly affects the results of geometry of weld beads of the final product. The…”

Section: Resultsmentioning

confidence: 99%

Thermal effects of AISI 304-L stainless steel after laser irradiation

Hirata,

Santos,

Vicente

et al. 2023

J. Eng. Exact Sci.

View full text Add to dashboard Cite

Laser Directed Energy Deposition (L-DED) is one of the Additive Manufacturing (AM) techniques that has emerged in recent years, and it is one of the disruptive technologies of the Industry 4.0. The selection of welding parameters of the L-DED system is critical for obtaining the desired results in metal deposition, however, due to its intrinsic nonlinear behavior in the process, the suitable choices need to be investigated. In this work it is presented an experimental setup to study the thermal effects on the metal substrate during the process of irradiation. With 40% and 60% of laser beam power, the irradiation results showed only a change in color, but no change in roughness or melting pool was formed. However, with 68% of laser power, a considerable increase of the substrate average temperature with values of more than 100°C at each adjacent track created was observed. The heat affected zone (HAZ) in this sample is expanded in both length and width as the temperature rises due the accumulation of heat during the irradiation process. It is concluded that the temperature of the substrate and other related process variables to be investigated must be controlled to achieve the desired bead formation.

show abstract

“…Thus, adopting different filling algorithms for interior and edges to achieve uniform interior filling and smooth edge-to-interior transitions will emerge as another research focus for EFed AM technologies. Additionally, integrating and coupling artificial intelligence and machine learning in the path planning of EFed AM processes also signals a promising future research direction [234].…”

Section: Rapid Prototyping Of Aviation Partsmentioning

confidence: 99%

Performance-control-orientated hybrid metal additive manufacturing technologies: state of the art, challenges, and future trends

Lv,

Liang,

et al. 2024

Int. J. Extrem. Manuf.

View full text Add to dashboard Cite

Metal additive manufacturing (AM) technologies have made significant progress in the basic theoretical field since their invention in the 1970s. However, performance instability during continuous processing, such as thermal history, residual stress accumulation, and columnar grain epitaxial growth, consistently hinders their broad application in standardized industrial production. To overcome these challenges, performance-control-oriented hybrid AM (HAM) technologies have been introduced. These technologies, by leveraging external auxiliary processes, aim to regulate microstructural evolution and mechanical properties during metal AM. In this context, HAM technologies for molten pool regulation and solidified material deformation are identified as energy field-assisted AM (EFed AM, e.g., ultrasonic, electromagnetic, heat, etc.) technologies and interlayer plastic deformation-assisted AM (IPDed AM, e.g., laser shock peening, rolling, ultrasonic peening, friction stir process, etc.) technologies, respectively. A detailed and systematic review of performance-control-oriented HAM technologies is then provided. This review covers the influence of external energy fields on the melting, flow, and solidification behavior of materials, and the regulatory effects of interlayer plastic deformation on grain refinement, nucleation, and recrystallization. Furthermore, the role of performance-control-oriented HAM technologies in managing residual stress conversion, metallurgical defect closure, mechanical property improvement, and anisotropy regulation is thoroughly reviewed and discussed. The review concludes with an analysis of future development trends in EFed AM and IPDed AM technologies.

show abstract

Research and application of artificial intelligence techniques for wire arc additive manufacturing: a state-of-the-art review

Cited by 35 publications

References 110 publications

Anomaly Detection in WAAM Deposition of Nickel Alloys—Single-Material and Cross-Material Analysis

Anomaly Detection in WAAM Deposition of Nickel Alloys—Single-Material and Cross-Material Analysis

Thermal effects of AISI 304-L stainless steel after laser irradiation

Performance-control-orientated hybrid metal additive manufacturing technologies: state of the art, challenges, and future trends

Contact Info

Product

Resources

About