Artificial intelligence (AI) based deep learning of defects, surface roughness and their linkage to mechanical performance of additively manufactured (AM) aluminum alloys

基于人工智能 (AI) 的缺陷、表面粗糙度及其与增材制造 (AM) 铝合金机械性能的联系的深度学习

基本信息

批准号：
549214-2019
负责人：
Inal, Kaan
金额：
$ 2.19万
依托单位：
University of Waterloo
依托单位国家：
加拿大
项目类别：
Alliance Grants
财政年份：
2020
资助国家：
加拿大
起止时间：
2020-01-01 至 2021-12-31
项目状态：
已结题

来源：
https://www.nserc-crsng.gc.ca/ase-oro/Details-Detailles_eng.asp?id=701078
关键词：
Artificial intelligence AI based deep

项目摘要

This Alliance project focuses on application of artificial intelligence (AI) based machine learning (ML) approaches to the field of additive manufacturing (AM) to develop a framework for bridging length scales and establishing linkages between process parameters, microstructural features and the resulting mechanical performance characteristics of additively manufactured parts. One of the major concerns in AM build parts is the variation in the microstructure along with the process induced defects such as porosity and surface roughness. The mechanical properties such as fatigue life of these materials are strongly governed by the choice of process parameters employed during the fabrication process. A wide variety of synergistic factors such as laser process parameters, surface roughness, residual stresses and porosity can significantly influence the fatigue behavior of AM parts. A lack of understanding of the linkages between such factors is the most critical decision-making gap for AM process optimization and its wide scale applicability. In these regards, artificial intelligence (AI) based machine learning tools are key enablers for identification and quantification of such essential material and process parameters or salient features for improved mechanical performance of AM parts. This project combines the latest advancements in the field of artificial intelligence and additive manufacturing, both theoretically and experimentally with an aim to investigate microstructure, property and performance relationships for 3D printed aluminum alloys by means of laser powder bed fusion (LPBF) technology.

该联盟项目的重点是基于人工智能（AI）的机器学习（ML）的应用方法，以在添加剂制造（AM）领域（AM）开发一个框架，以桥接长度尺度，并在过程参数，微观结构特征和由此产生的添加零件的机械性能特征之间建立联系。 AM构建部分的主要关注点之一是微观结构的变化以及过程引起的缺陷，例如孔隙度和表面粗糙度。这些材料的疲劳寿命等机械性能受到制造过程中采用的过程参数的选择。各种各样的协同因素，例如激光过程参数，表面粗糙度，残留应力和孔隙率可以显着影响AM部分的疲劳行为。缺乏对这些因素之间联系的了解是AM过程优化及其广泛适用性的最关键决策差距。在这些方面，基于人工智能（AI）的机器学习工具是识别和量化此类基本材料和过程参数或显着特征以改善AM零件机械性能的关键推动者。该项目结合了人工智能领域和添加剂制造领域的最新进步，无论是在理论上还是实验上，旨在通过激光粉末融合（LPBF）技术研究3D印刷铝合金的微观结构，财产和性能关系。