Investigation of a path planning solution for wire + arc additive manufacture.

Date

2018-12

Journal Title

Journal ISSN

Volume Title

Publisher

Cranfield University

Department

SATM

Type

Thesis or dissertation

ISSN

Format

Free to read from

Citation

Abstract

Wire + Arc Additive Manufacturing (WAAM) has become a crucial asset for industrial manufacturing in the field of medium to large metallic deposition thanks to its high-rate deposition of various metals, its low-cost equipment and a potentially unlimited build volume. A key element for commercial deployment is to develop an intuitive path planning software, which can determine the optimal deposition strategy, whilst respecting WAAM’s constraints inherent to arc welding deposition. Traditional approaches to additive manufacturing path planning are often derived from CNC machining, but these strategies are incompatible with some fundamental characteristics of WAAM. For this reason, the present work aims to investigate a path planning solution entirely focused on the WAAM requirements. The architecture of a Path Generator Framework for WAAM is, thus, first introduced to offer complete freedom of path planning development all along this study. To validate the developed framework, a feature- based approach is presented: this allows the fast and efficient deployment of the WAAM technology for a limited range of geometric features and sets up the basis of path planning for WAAM. Then, a more flexible solution called Modular Path Planning is introduced to incorporate the modularity of feature-based design into the traditional layer-by-layer build strategy. By assisting the user in dividing each layer into individual deposition sections, this method enables users to adapt the path strategy to the targeted geometry allowing the construction of a wide variety of complex geometries. Finally, a deep learning solution called DeepWAAM is proposed to reach, in the future, a fully automated path planning solution for WAAM by automatically dividing build layers into deposition sections with no need for user intervention.

Description

Software Description

Software Language

Github

Keywords

WAAM, wire and arc additive manufacturing, path planning, toolpath generation, robotics simulation framework, WAAM platform, deep learning, machine learning, neural network, AI

DOI

Rights

© Cranfield University, 2018. All rights reserved. No part of this publication may be reproduced without the written permission of the copyright holder.

Relationships

Relationships

Supplements

Funder/s