Life Prediction for Directed Energy Deposition‐Manufactured 316L Stainless Steel using a Coupled Crystal Plasticity–Machine Learning Framework
Life Prediction for Directed Energy Deposition‐Manufactured 316L Stainless Steel using a Coupled Crystal Plasticity–Machine Learning Framework
复制标题
定向能量沉积的寿命预测——使用耦合晶体塑性制造的 316L 不锈钢——机器学习框架
DOI：
10.1002/adem.202201429
复制
发表时间：
2023
影响因子：
3.6
通讯作者：
Mushongera, Leslie T.
中科院分区：
文献类型：
--
作者：
Ye, Wenye;Zhang, Xing;Hohl, Jake;Liao, Yiliang;Mushongera, Leslie T.
增材制造的不锈钢由于其理想的性能而变得越来越受欢迎，但它们在结构部件中的机械性能尚未完全了解。具体来说，柱状微观结构对疲劳行为的影响仍不清楚。典型的定向能量沉积 (DED) 制造的 316L 不锈钢微观结构由具有等轴晶粒和柱状晶粒的不同区域组成。为了回答 DED 制造的 316L 不锈钢微观结构的这些区域如何单独影响局部机械行为（例如疲劳强度、应力/应变分布和疲劳寿命）的问题，进行晶体塑性模拟来研究微观结构对局部机械行为（例如疲劳强度、应力/应变分布和疲劳寿命）的影响。模拟发现，当载荷平行于柱状晶的长轴时，柱状组织比等轴组织表现出更好的疲劳强度，但当载荷垂直时，强度下降。这项研究还使用机器学习来预测疲劳寿命，这与晶体塑性模型具有良好的一致性。该研究表明，晶体塑性与机器学习相结合的方法是预测增材制造部件疲劳行为的有效方法。
Additively manufactured stainless steels have become increasingly popular due to their desirable properties, but their mechanical behavior in structural parts is not yet fully understood. Specifically, the impact of columnar microstructures on fatigue behavior is still unclear. A typical directed energy deposition (DED)‐fabricated 316L stainless steel microstructure consists of distinct zones with equiaxed and columnar grains. To answer the question of how these zones of a DED‐fabricated 316L stainless steel microstructure affect the local mechanical behavior individually, such as the fatigue strength, stress/strain distribution, and fatigue life, crystal plasticity simulations are conducted to investigate the influence of microstructure on local mechanical behavior such as fatigue strength, stress/strain distribution, and fatigue life. The simulations find that columnar microstructures exhibit better fatigue strength than equiaxed structures when the load is parallel to the major axis of the columnar grains, but the strength decreases when the load is perpendicular. This study also uses machine learning to predict fatigue life, which shows good agreement with crystal plasticity modeling. The study suggests that the combined crystal plasticity–machine learning approach is an effective way to predict the fatigue behavior of additively manufactured components.
DOI：
--
发表时间：
2001
期刊：
影响因子：
--
作者：
R. McGinty
通讯作者：
R. McGinty