A Data Analytics Approach to Discovering Unique Microstructural Configurations Susceptible to Fatigue

被引：5

作者：

Jha, S. K. ^{[2
]}

Brockman, R. A. ^{[2
]}

Hoffman, R. M. ^{[2
]}

Sinha, V. ^{[3
]}

Pilchak, A. L. ^{[1
]}

Porter, W. J., III ^{[2
]}

Buchanan, D. J. ^{[2
]}

Larsen, J. M. ^{[1
]}

John, R. ^{[1
]}

机构：

[1] US Air Force, Res Lab, Mat & Mfg Directorate, RXCM, Wright Patterson AFB, OH 45433 USA

[2] Univ Dayton, Res Inst, 300 Coll Pk, Dayton, OH 45469 USA

[3] UES Inc, 4401 Dayton Xenia Rd, Dayton, OH 45432 USA

来源：

JOM | 2018年 / 70卷 / 07期

关键词：

CRACK-INITIATION; TITANIUM-ALLOY; HIGH-CYCLE; DATA SET; CLUSTERS; NUMBER; LIFE;

D O I：

10.1007/s11837-018-2881-5

中图分类号：

T [工业技术];

学科分类号：

08 ;

摘要：

Principal component analysis and fuzzy c-means clustering algorithms were applied to slip-induced strain and geometric metric data in an attempt to discover unique microstructural configurations and their frequencies of occurrence in statistically representative instantiations of a titanium alloy microstructure. Grain-averaged fatigue indicator parameters were calculated for the same instantiation. The fatigue indicator parameters strongly correlated with the spatial location of the microstructural configurations in the principal components space. The fuzzy c-means clustering method identified clusters of data that varied in terms of their average fatigue indicator parameters. Furthermore, the number of points in each cluster was inversely correlated to the average fatigue indicator parameter. This analysis demonstrates that data-driven methods have significant potential for providing unbiased determination of unique microstructural configurations and their frequencies of occurrence in a given volume from the point of view of strain localization and fatigue crack initiation.

引用

页码：1147 / 1153

页数：7

共 33 条

[11] Hartigan J. A., 1979, Applied Statistics, V28, P100, DOI 10.2307/2346830
[12] A geometric approach to modeling microstructurally small fatigue crack formation: III. Development of a semi-empirical model for nucleation
Hochhalter, J. D.
Littlewood, D. J.
Veilleux, M. G.
Bozek, J. E.
Maniatty, A. M.
Rollett, A. D.
Ingraffea, A. R.
[J]. MODELLING AND SIMULATION IN MATERIALS SCIENCE AND ENGINEERING, 2011, 19 (03)
[13] Jang R., 2002, DATA SET CLUSTERING
[14] Jha S. K., 2018, UNPUB
[15] Deformation heterogeneities and their role in life-limiting fatigue failures in a two-phase titanium alloy
Jha, Sushant K.
Szczepanski, Christopher J.
John, Reji
Larsen, James M.
[J]. ACTA MATERIALIA, 2015, 82 : 378 - 395
[16] Characterization of fatigue crack-initiation facets in relation to lifetime variability in Ti-6Al-4V
Jha, Sushant K.
Szczepanski, Christopher J.
Golden, Patrick J.
Porter, William J., III
John, Reji
[J]. INTERNATIONAL JOURNAL OF FATIGUE, 2012, 42 : 248 - 257
[17] Jolliffe I T., 2002, Principal Component Analysis, P1, DOI [DOI 10.1007/0-387-22440-8_1, DOI 10.1007/B98835]
[18] Kaufman L, 2009, FINDING GROUPS DATA
[19] Reducing uncertainty in fatigue life limits of turbine engine alloys
Larsen, J. M.
Jha, S. K.
Szczepanski, C. J.
Caton, M. J.
John, R.
Rosenberger, A. H.
Buchanan, D. J.
Golden, P. J.
Jira, J. R.
[J]. INTERNATIONAL JOURNAL OF FATIGUE, 2013, 57 : 103 - 112
[20] Machine learning applications in genetics and genomics
Libbrecht, Maxwell W.
Noble, William Stafford
[J]. NATURE REVIEWS GENETICS, 2015, 16 (06) : 321 - 332

← 1 2 3 4 →