Foundations of Intelligent Additive Manufacturing

Abstract

During the last decade, additive manufacturing has become increasingly popular for rapid prototyping, but has remained relatively marginal beyond the scope of prototyping when it comes to applications with tight tolerance specifications, such as in aerospace. Despite a strong desire to supplant many aerospace structures with printed builds, additive manufacturing has largely remained limited to prototyping, tooling, fixtures, and non-critical components. There are numerous fundamental challenges inherent to additive processing to be addressed before this promise is realized. One ubiquitous challenge across all AM motifs is to develop processing-property relationships through precise, in situ monitoring coupled with formal methods and feedback control. We suggest a significant component of this vision is a set of semantic layers within 3D printing files relevant to the desired material specifications. This semantic layer provides the feedback laws of the control system, which then evaluates the component during processing and intelligently evolves the build parameters within boundaries defined by semantic specifications. This evaluation and correction loop requires on-the-fly coupling of finite element analysis and topology optimization. The required parameters for this analysis are all extracted from the semantic layer and can be modified in situ to satisfy the global specifications. Therefore, the representation of what is printed changes during the printing process to compensate for eventual imprecision or drift arising during the manufacturing process.

Machine Learning and Data Analytics for Design and Manufacturing of High-Entropy Materials Exhibiting Mechanical or Fatigue Properties of Interest

“…Other related work specifically addressing the prediction of properties of additively manufactured components includes [198][199][200]. For relevant papers on HEA design for additive manufacturing, refer to [201][202][203].…”

Open source computer vision-based layer-wise 3D printing analysis

“…Okaro et al [47] introduced a semi-supervised machine learning algorithm based on large sets of photodiode data for automatic "faulty" and "acceptable" tensile strength assessment in laser power bed fusion additive manufacturing. Garanger et al [48] suggested a number of semantic rules within 3-D printing files, which provide desired specifications and, based on material properties, real-time topology and finite element analysis, generate feedback laws for the control system. Yuan et al [49] developed a two-step machine learning approach to real-time laser track welds assessment in LPBF processes.…”