Machine learning integrated design for additive manufacturing

  • PDF / 2,400,431 Bytes
  • 14 Pages / 595.276 x 790.866 pts Page_size
  • 94 Downloads / 286 Views

DOWNLOAD

REPORT


Machine learning integrated design for additive manufacturing Jingchao Jiang1

· Yi Xiong1

· Zhiyuan Zhang1 · David W. Rosen1,2

Received: 3 July 2020 / Accepted: 11 November 2020 © Springer Science+Business Media, LLC, part of Springer Nature 2020

Abstract For improving manufacturing efficiency and minimizing costs, design for additive manufacturing (AM) has been accordingly proposed. The existing design for AM methods are mainly surrogate model based. Due to the increasingly available data nowadays, machine learning (ML) has been applied to medical diagnosis, image processing, prediction, classification, learning association, etc. A variety of studies have also been carried out to use machine learning for optimizing the process parameters of AM with corresponding objectives. In this paper, a ML integrated design for AM framework is proposed, which takes advantage of ML that can learn the complex relationships between the design and performance spaces. Furthermore, the primary advantage of ML over other surrogate modelling methods is the capability to model input–output relationships in both directions. That is, a deep neural network can model property–structure relationships, given structure–property input–output data. A case study was carried out to demonstrate the effectiveness of using ML to design a customized ankle brace that has a tunable mechanical performance with tailored stiffness. Keywords Additive manufacturing · Design for AM · Machine learning

Introduction Additive Manufacturing (AM) fabricates products from digital 3D models in a piece-by-piece, line-by-line or layer-bylayer manner, which is different from conventional manufacturing processes (Jiang et al. 2019a, b, c, d, e; Wei et al. 2020; Xiong et al. 2020). This gives AM more freedom on design as AM can fabricate more complicated parts, theoretically in any shape, without spending more efforts on manufacturing processes. AM bonds, places, and/or transforms volumetric primitives or elements (voxels) of raw material to fabricate the final products. The size and shape of each voxel and the bonding strength between voxels depend on the raw material properties, the AM machine (e.g., nozzle diameter), and the process parameters (e.g., print temperature, print speed, laser power). Therefore, design for additive manufacturing (DfAM) was proposed with the aim of designing and optimizing the product together with its manufacturing systems to increase the product’s quality and performance, and to

B

David W. Rosen [email protected]

1

Digital Manufacturing and Design Center, Singapore University of Technology and Design, Singapore, Singapore

2

The G. W. Woodruff School of Mechanical Engineering, Georgia Institute of Technology, Atlanta, GA 30332, USA

minimize the development time and cost (Gibson et al. 2010; Huang et al. 2020; Kim et al. 2019). DfAM actually is a type of Design for Manufacturing and Assembly (DfMA) but is quite different from traditional DfMA. The unique capabilities of AM technologies make designers re-think the traditiona