不锈钢材料微观结构的数据约束模型表征  

Microstructure Characterization of a Stainless-Steel Sample with Data-constrained Modelling

作　　者：杨玉双 王海鹏[2] 王晋庥 王润花 李建莉 刘锐 YANG Yushuang;WANG Haipeng;WANG Jinxiu;WANG Runhua;LI Jianli;LIU Rui(Commonwealth Scientific and Industrial Research Organisation,VIC3168,Australia;College of Physics and Elecronics Engineering,Shanxi Universiy,Taiyuan 030006,China;Institute of Theoretical Plysics,Shanxi University,Taiyuan 030006,China)

机构地区：[1]澳大利亚联邦科学与工业研究组织,澳大利亚维多利亚州3168 [2]山西大学物理电子工程学院,山西太原030006 [3]山西大学理论物理研究所,山西太原030006

出　　处：《山西大学学报（自然科学版）》2022年第3期693-700,共8页Journal of Shanxi University(Natural Science Edition)

摘　　要：不锈钢材料中的微观缺陷分布对其机械性能及耐蚀性具有重要影响。文章对一316L不锈钢样品进行了X光断层扫描成像(X-ray CT),发展了一种CT图像移动体积平均方法以去除线性吸收系数空间不均匀性伪影。建立了样品的数据约束模型,提取到样品中包含小于CT体元尺寸的缺陷分布。基于计算所得每个体元中缺陷的体积分数对缺陷的联通团簇进行了计算,以此为基础估算了样品中的缺陷分布尺寸范围。研究结果表明,相比目前主流的图像阈值分割算法,数据约束模型可有效提取样品中小于CT体元尺寸的缺陷分布,为不锈钢样品的三维微观结构表征提供了新的思路。The defect distribution in stainless-steel materials has significant impacts on its mechanical properties and corrosion resistance.In this paper,a commercial 316L stainless-steel sample was imaged with X-ray computed tomography(X-ray CT).A moving-volume averaging algorithm was developed to eliminate the spatial inconsistency in X-ray CT reconstructed linear-absorption coefficients.Microscopic defects in the sample were characterized with a data-constrained modelling(DCM)approach,including those defects which were smaller than the X-ray CT voxel size.Using the DCM computed volume fractions of defects in each voxel,the connected defect clusters were calculated and the size range of defects was estimated.The results show that the DCM approach can effectively detect defects which are smaller than the X-ray CT voxels.Given that those small defects would be missed out with the current mainstream image thresholding segmentation method,the DCM approach opens up a new avenue in three-dimensional microstructure characterization of stainless-steel materials.

关 键 词：数据约束模型 316L不锈钢 X射线CT成像 无损检测 

分 类 号：O242[理学—计算数学]