增材制造材料缺陷表征及结构完整性评定方法研究综述

金属增材制造过程中不可避免地会产生气孔和未熔合缺陷。尽管采取参数优化和后热处理能够在一定程度上降低缺陷水平，但至今尚无有效方法予以完全消除。这些缺陷作为典型的应力集中源，会诱导疲劳裂纹形核，从而大幅降低材料的疲劳强度和寿命，被视为增材构件可靠性服役的“顽疾”。从静态缺陷表征、动态缺陷演化、缺陷分级、缺陷-疲劳强度设计方法以及缺陷-疲劳寿命评估技术等五个方面论述增材制造缺陷与疲劳行为的研究进展。重点介绍借助X射线成像技术开展缺陷特征及演化的三维、无损、可视化表征与定量统计方法；进一步地，论述基于同步辐射光源的原位力学和疲劳测试系统及表征方法及其在原位、无损、实时、动态追踪缺陷或者裂纹演化机制方面中的应用；增材缺陷具有全域分布、形态多样、尺寸跨度大等特征，总结六种缺陷等级判断方法；在缺陷容限和损伤容限框架内，建立基于材料表面/亚表面/内部缺陷特征的疲劳强度和寿命评价方法。最后，指出借助数据驱动的高通量试验平台和机器学习算法、多尺度多物理场数值模拟是实现增材制造材料工艺设计-缺陷表征-性能评价一体化研究的重要研究课题。

Review on Defect Characterization and Structural Integrity Assessment Method of Additively Manufactured Materials

Manufacturing defects have been an unavoidable feature of additive manufacturing (AM) processed metals:typically these comprise gas pores and lack of fusion defects. The applications of optimized process parameters and post-AM heat treatment are able to reduce these defects to a certain degree. Unfortunately, to date there is no effective way to completely eliminate them. Defects can have a detrimental effect on the fatigue strength and life of a material since they act as potential crack initiation sites due to high stress concentration, posing a significant threat to the structural integrity of AM processed components. The research progress of AM defect behavior is summarized from five aspects:static defect characterization, dynamic defect evolution, defect classification, defect-fatigue strength design and defect-fatigue life evaluation. First, the inherent manufacturing defects induced by AM are characterized and quantified using X-ray computed microtomography in a three-dimensional and non-destructive way. Special studies on the spatial defect or crack evolution behavior are also reviewed by using a novel in situ synchrotron X-ray computed microtomography during cyclic loading in an in situ, real-time and dynamic way. AM defects are characterized by global distribution, diverse morphologies and large size spans. Six ranking strategies, having varying levels of complexity, are proposed to estimate the threat posed by different defects. Within the framework of defect tolerance and damage tolerance, some methods are developed to evaluate the fatigue strength and lifetime in terms of the defect geometry at the surface, subsurface, and in the interior of the materials. Finally, it is pointed out that the data-driven high-throughput testing approach and machine learning algorithms as well as the multi-scale & multi-physics numerical simulation are of vital significance for the integrated exploration on the process design-defect characterization-performance evaluation of AM processed metals.