Experimental Investigation of Stochastic Jumps during Crack Initiation and Growth in IN718
Experimental Investigation of Stochastic Jumps during Crack Initiation and Growth in IN718
复制标题
DOI：
--
复制
发表时间：
2019-06
期刊：
影响因子：
--
通讯作者：
Joel Lindsay;S. Papanikolaou;T. Musho
中科院分区：
文献类型：
--
作者：
Joel Lindsay;S. Papanikolaou;T. Musho
他的研究调查了几种不同加载条件下Inconel 718 （IN718）裂纹跳跃噪声的统计意义。采用直流电位降（DCPD）法对裂纹长度进行了原位测量。在R=0.15时收集六个不同峰值负载的数据，以统计上显著的路径数量。有限元导出的校准曲线将测量电位与裂纹长度联系起来。在后续循环中，裂纹长度跳变的均值随着载荷的增加而增加，裂纹长度跳变分布的范围随着载荷的增加而减小，而噪声具有非零的均值分布。这项研究的结果表明，裂纹长度跳跃不是随机事件，而是包含统计特征，可以与机器学习方法一起使用，以更好地理解ni基高温合金的疲劳进展。
his study investigates the statistical significance of crack jump noise in Inconel 718 (IN718) for several different loading conditions. A direct current potential drop (DCPD) method is used to experimentally measure in-situ the crack length. Data is collected for six different peak loads at R=0.15 for a statistically significant number of trails. FEA-derived calibration curves relate measured potential to crack length. We determine that the mean crack length jumps, over subsequent cycles, increased with loading, the range of the crack length jump distributions decreases with increasing load, while the noise has a non-zero mean distribution. Findings from this study suggest that crack length jumps are not random events but contain statistical features that can potentially be used with machine learning approaches to better understand fatigue progression in Ni-based superalloys.