Directed energy deposition processes and process design by artificial intelligence

Rafael Paiotti Marcondes Guimaraes*, Florian Pixner, N Enzinger, Carlos Alberto Belei Feliciano, Pedro Effertz, S. T. Amancio-Filho

*Corresponding author for this work

Research output: Chapter in Book/Report/Conference proceedingChapterpeer-review

Abstract

Differently from the powder-based additive (AM) techniques, wire-based ones aim to produce larger parts with moderate complexity and semi-finished features. For this reason, wire-based additive manufacturing (WBAM) is a strong candidate to compete with subtractive manufacturing techniques and forging. Compared to powder-based, WBAM presents higher layer thickness, thus having higher build-up and deposition rates, despite lower-dimensional accuracies and surface roughness. However, WBAM and powder-based AM are not concurrent technologies since both produce parts for different applications. Problems in EBAM are amplified due to the large part sizes; residual stresses and distortion from the high heat input directly impair the part quality. Furthermore, the thermal history resulting from the repetitive layering is responsible for generating complex metallurgical phenomena, requiring investigations to select the best combination of process parameters and deposition strategy for each material. Nonetheless, since WBAM processes use the same principles of welding with filler material (or, in the case of AM, feedstock), advances are taking place faster. Accordingly, WBAM techniques such as wire and arc AM (WAAM), wire-based electron beam AM (w-EBAM or EBAM), and wire and laser additive manufacturing (WLAM) have achieved maturity, proving their feasibility in producing parts for critical structural applications in fields such as aerospace. This chapter is destined to expose recent research conducted on the WBAM of Ti and NiTi alloys, compare different WBAM techniques, introduce the use of cored wires for AM, present new AM processes, and give a brief and concise overview of artificial intelligence applied to additive manufacturing. First, the section clarifies cold metal transfer (CMT) as an AM technique presenting some of its features and capabilities, subsequently comparing it to WAAM. This section is finished with a comparison between solid and cored wires, the latter a new need for fusion wire-based processes; pros and cons are given based on practical observations. The second section groups two classes of alloys additively manufactured by EBAM: Ti and NiTi alloys. EBAM is firstly introduced where general features of the process are presented, followed by process parameters, pros and cons. In sequence, the EBAM of Ti-64 alloy is shown with results based on microstructural aspects, thermal history, and phase transformation. The EBAM of NiTi follows a different approach, exploring recent publications of this newborn processing of NiTi alloys. Comparisons with techniques (WAAM and LPBF) show how EBAM may influence functional properties. Therefore, the focus is put on the influence of processing on the transformation temperatures and superelastic behaviour. The outlook comprises two wire-based AM techniques: resistance welding and liquid metal AM. The former recently introduced and baptised Joule Printing gained visibility due to its astonishing printing speed. Liquid metal was introduced two decades ago and is based on the Lorentz force to drop the feedstock. General information is exposed to understand each process. Another important segment of AM bases its layering method on friction, specifically friction stir welding. Features of this process are indicated, and practical examples are depicted with Al and Ni alloys. Ultrasonic metal AM is also represented afterwards, with basic information about the process and examples of components. At last, the effects of coupling artificial intelligence to additive manufacturing and its benefits are enlightened, with applications for both machine and deep learning as well as future trends in, e.g., topology optimisation.

Original languageEnglish
Title of host publicationAdvances in Metal Additive Manufacturing
PublisherWoodhead Publishing
Pages105-146
Number of pages42
Edition1
ISBN (Electronic)9780323914680
ISBN (Print)9780323912303
DOIs
Publication statusPublished - 1 Jan 2022

Publication series

NameWoodhead Publishing Reviews: Mechanical Engineering Series

Keywords

  • artificial intelligence
  • cold metal transfer
  • friction-based additive manufacturing
  • joule printing
  • metal-cored wires
  • shape memory alloys
  • titanium alloys
  • ultrasonic metal additive manufacturing
  • wire-arc additive manufacturing
  • wire-based electron beam additive manufacturing

ASJC Scopus subject areas

  • Engineering(all)

Fields of Expertise

  • Advanced Materials Science

Fingerprint

Dive into the research topics of 'Directed energy deposition processes and process design by artificial intelligence'. Together they form a unique fingerprint.

Cite this