In the present study, manganese sulfide (MnS) inclusions in the high-strength steel were observed by mainly three observation methods (optical microscope, ultrasonic test and serial sectioning) to characterize the size, location and shape distributions across multiple length scales. For the inclusion size, ultrasonic C-scan imaging and three-dimensional internal structure observation with serial sectioning were used to measure the distributions of the square root of the projected area of the inclusion. The obtained size distributions were combined by setting the threshold of ultrasonic amplitude. The validity of the amplitude threshold was verified by observing several inclusions with X-ray CT. The spatial distributions of inclusions were also obtained by the three observation methods, and analyzed on the basis of the coefficient of variation of the mean near-neighbor distance of inclusions (COV_d). The results of analyzing COV_d in both 2D and 3D spaces revealed that the inclusions in this material were arranged in clusters. For the inclusion shape, the three-dimensional geometries of inclusions were reconstructed from the images obtained by the serial sectioning method, and simplified to ellipsoid by principal component analysis. From the above results, the distributions of inclusion size, aspect ratio and direction (angle between rolling direction and major axis) were successfully obtained. The inclusion distributions were applied to fatigue prediction model, and the fatigue crack initiation life and total fatigue life of the high-strength steel were calculated. The calculation results showed that the multiscale analysis of inclusions would be useful for fatigue life prediction.

在本研究中，主要通过三种观察方法（光学显微镜，超声测试和连续切片）观察了高强度钢中的硫化锰（MnS）夹杂物，以表征多个长度尺度上的尺寸，位置和形状分布。对于夹杂物的尺寸，使用超声C扫描成像和连续切片的三维内部结构观察来测量夹杂物投影区域的平方根分布。通过设置超声振幅的阈值来合并获得的尺寸分布。通过使用X射线CT观察几个夹杂物，可以验证振幅阈值的有效性。还通过三种观察方法获得了夹杂物的空间分布：_d）。在2D和3D空间中分析COV _d的结果表明，这种材料中的夹杂物呈簇状排列。对于夹杂物的形状，从通过连续切片方法获得的图像中重建夹杂物的三维几何形状，并通过主成分分析将其简化为椭圆体。根据以上结果，成功地获得了夹杂物尺寸，长径比和方向（轧制方向与主轴线之间的夹角）的分布。将夹杂物分布应用于疲劳预测模型，计算出高强度钢的疲劳裂纹萌生寿命和总疲劳寿命。计算结果表明，夹杂物的多尺度分析将有助于疲劳寿命的预测。