Metal additive manufacturing(AM)offers flexibility and cost-effectiveness for printing complex parts but is limited to few alloys.Qualifying new alloys requires process parameter optimisation to produce consistent,hig...Metal additive manufacturing(AM)offers flexibility and cost-effectiveness for printing complex parts but is limited to few alloys.Qualifying new alloys requires process parameter optimisation to produce consistent,high-quality components.High-resolution X-ray computed tomography(XCT)has not been effective for this task due to artifacts,slow scan speed,and costs.We propose a deep learning-based approach for rapid XCT acquisition and reconstruction of metal AM parts,leveraging computer-aided design models and physics-based simulations of nonlinear interactions between X-ray radiation and metals.This significantly reduces beam hardening and common XCT artifacts.We demonstrate high-throughput characterisation of over a hundred AlCe alloy components,quantifying improvements in characterisation time and quality compared to high-resolution microscopy and pycnometry.Our approach facilitates investigating the impact of process parameters and their geometry dependence in metal AM.展开更多
基金This manuscript has been authored by UT-Battelle,LLC,under contract DE-AC05-00OR22725 with the US Department of Energy(DOE)Research sponsored by the US Department of Energy,Office of Energy Efficiency and Renewable Energy,Advanced Manufacturing Office and Technology Commercialisation Fund(TCF-21-24881)+1 种基金under contract DE-AC05-00OR22725 with UT-Battelle,LLCThe US government retains and the publisher,by accepting the article for publication,acknowledges that the US government retains a nonexclusive,paid-up,irrevocable,worldwide license to publish or reproduce the published form of this manuscript,or allow others to do so,for US government purposes.DOE will provide public access to these results of federally sponsored research in accordance with the DOE Public Access Plan(http://energy.gov/downloads/doe-public-access-plan).
文摘Metal additive manufacturing(AM)offers flexibility and cost-effectiveness for printing complex parts but is limited to few alloys.Qualifying new alloys requires process parameter optimisation to produce consistent,high-quality components.High-resolution X-ray computed tomography(XCT)has not been effective for this task due to artifacts,slow scan speed,and costs.We propose a deep learning-based approach for rapid XCT acquisition and reconstruction of metal AM parts,leveraging computer-aided design models and physics-based simulations of nonlinear interactions between X-ray radiation and metals.This significantly reduces beam hardening and common XCT artifacts.We demonstrate high-throughput characterisation of over a hundred AlCe alloy components,quantifying improvements in characterisation time and quality compared to high-resolution microscopy and pycnometry.Our approach facilitates investigating the impact of process parameters and their geometry dependence in metal AM.
