Knowledge-based manufacturability assessment for optimization of additive manufacturing processes based on automated feature recognition from CAD models

Abstract

Additive manufacturing (AM) is a significant development of the manufacturing sector that has emerged during the last decades and tends to change the way products are designed, manufactured, and repaired, enabling unprecedented flexibility levels. The unique process mechanism of AM enables the realization of complex designs after considering design limitations, which are unique to each process mechanism and machine. These limitations, together with the lack of established AM-related knowledge, lead to the design of parts that are not fully conforming with AM buildability restrictions, resulting in failed builds. To this end, this work presents a methodology that enables to embed the AM-related knowledge and use it for an automated manufacturability assessment. The 3D model of a part is used as an input in an.stp format, and the features that are relevant for AM manufacturability are recognized from the global CAD. Then, an analysis of the manufacturability of these features according to the limitations of the process and/or machine is performed, and design changes are suggested to the user aiming to improve the process outputs. The whole methodology is implemented in a software tool with an intuitive user-interface that supports the users in the design of parts that will be made with additive manufacturing.

Design for additive manufacturing of topology-optimized structures based on deep learning and transfer learning

“…These geometric restrictions involve filleting sharp corners and increasing the thickness of thin walls to reduce stress concentrations and residual stresses. Recently, many studies have been conducted to analyze and improve the manufacturability of AM parts (Shi et al , 2018; Lianos et al , 2020; Mayerhofer et al , 2021; Stavropoulos et al , 2022). For instance, Stavropoulos et al (2022) used 3D models for feature recognition based on CAD software.…”

“…Recently, many studies have been conducted to analyze and improve the manufacturability of AM parts (Shi et al , 2018; Lianos et al , 2020; Mayerhofer et al , 2021; Stavropoulos et al , 2022). For instance, Stavropoulos et al (2022) used 3D models for feature recognition based on CAD software. They aimed to extract critical areas, including overhangs and unsupported areas, wall thicknesses, aspect ratios and holes, and they investigated the manufacturability based on them and DfAM guidelines for the FDM process.…”

A Novel Feature-Based Manufacturability Assessment System for Evaluating Selective Laser Melting and Subtractive Manufacturing Injection Moulding Tool Inserts

“…The geometric features of the mould core and cavity were evaluated for manufacturability using a fuzzy analytic hierarchy process (Fuzzy-AHP) methodology, as geometric compatibility for manufacturing a feature is characterised by a 'pass' or 'fail' approach. Another study proposed an AM manufacturability assessment approach that incorporates the use of automated feature recognition from the original CAD model with established process knowledge to optimise process outputs for metal AM processes [6].…”

Efficient GPU Cloud architectures for outsourcing high-performance processing to the Cloud

“…This infrastructure is being enrich by adding specific purpose computing devices, such as Graphic Processing Units (GPU). Their use is specially useful in external processing architectures where computing intensive tasks of specialized fields such as CAD/CAM [5,35] or Artificial Intelligence (AI) [14] are being offloaded to the Cloud, where the use of GPUs provides better performance [13,32], and allows commodity hardware to perform this operations. This brings superior parallel computing capabilities that tackles new computing intensive needs [34].…”