Knowledge assisted machine learning to clarify pore influence on fatigue life of forging/additive hybrid manufactured Ti-17 alloy

Abstract

Forging/additive hybrid manufactured Ti alloy parts suffer from relatively low fatigue life due to the existence of metallurgical defects in the transition zone, which also brings difficulty to fatigue life modeling. In this work, the synergistic effect of pore size and location on the rotating-bending fatigue life of hybrid manufactured Ti-5Al-2Sn-2Zr-4Mo-4Cr (Ti-17) samples was systematically investigated with the combination of machine learning approaches and physical knowledge. A machine learning framework with a back propagation neural network and generative adversarial network (GAN) was constructed and employed on sparse and limited datasets. A general and interpretable model was obtained with a high level of 90% confidence. In general, the fatigue life of hybrid manufactured Ti-17 alloys decreases with pore size and increases with its distance to surface. Specifically, critical sizes were obtained for near-surface and in-depth pores that have negligible influence on fatigue life of hybrid manufactured samples with respect to pore-free samples. The present work thus provides a systematic platform for the evaluation of the fatigue performance of hybrid manufactured titanium alloys.

Keywords

Hybrid manufacturing / fatigue / titanium / defect / machine learning

References

[1] Zhang D,Gibson MA.Additive manufacturing of ultrafine-grained high-strength titanium alloys.Nature2019;576:91-5

[2] Korkmaz ME,Garcia-collado A,Krolczyk GM.A technical overview of metallic parts in hybrid additive manufacturing industry.J Mater Res Technol2022;18:384-95

[3] Hirtler M,Sydow B,Bambach M.A study on the mechanical properties of hybrid parts manufactured by forging and wire arc additive manufacturing.Procedia Manuf2020;47:1141-8

[4] Murakami Y.Effects of defects, inclusions and inhomogeneities on fatigue strength.Int J Fatigue1994;16:163-82

[5] Murakami Y.Effects of small defects and nonmetallic inclusions on the fatigue strength of metals.JSME Int J Ser1989;32:167-80

[6] Piascik RS.Environmental fatigue of an Al-Li-Cu alloy: part I. Intrinsic crack propagation kinetics in hydrogenous environments.Metall Trans A1991;22:2415-28

[7] Piascik RS.Environmental fatigue of an Al-Li-Cu alloy: Part II. Microscopic hydrogen cracking processes.Metall Trans A1993;24:2751-62

[8] Davidson CJ,Machin AS.The effect of solution heat-treatment time on the fatigue properties of an Al-Si-Mg casting alloy.Fatigue Fract Eng Mat Struct2002;25:223-30

[9] Zhu M,Chen J.Influence of microstructure and microdefects on long-term fatigue behavior of a Cr–Mo–V steel.Mat Sci Eng A2012;546:90-6

[10] Zhu M,Du Y.Very high cycle fatigue behavior of a low strength welded joint at moderate temperature.Int J Fatigue2012;40:74-83

[11] Zhu M,Xuan F.Fatigue life and mechanistic modeling of interior micro-defect induced cracking in high cycle and very high cycle regimes.Acta Mater2018;157:259-75

[12] Jordan MI.Machine learning: trends, perspectives, and prospects.Science2015;349:255-60

[13] Bao H,Wu Z,Peng X.A machine-learning fatigue life prediction approach of additively manufactured metals.Eng Fract Mech2021;242:107508

[14] Zhan Z,Meng Q.Data-driven fatigue life prediction in additive manufactured titanium alloy: a damage mechanics based machine learning framework.Eng Fract Mech2021;252:107850

[15] Ye S,Li M.Deep learning method for determining the surface elastic moduli of microstructured solids.Extreme Mech Lett2021;44:101226

[16] Luo C,Keshtegar B,Branco R.Active Kriging-based conjugate first-order reliability method for highly efficient structural reliability analysis using resample strategy.Comput Method Appl M2024;423:116863

[17] Wu S,Wang Y,Zhang Z.Multiaxial low-cycle fatigue life model for notched specimens considering small sample characteristics.IJSI2024;15:948-73

[18] Yang S,Wang H.A novel learning function for adaptive surrogate-model-based reliability evaluation.Philos Trans A Math Phys Eng Sci2024;382:20220395

[19] Tarsha-Kurdi F,Grussenmeyer P. Model-driven and da-ta-driven approaches using LIDAR data: analysis and comparison. 2007. Available from: https://www.researchgate.net/publication/32227577_Model-driven_and_data-driven_approaches_using_LIDAR_data_Analysis_and_comparison. [Last accessed on 29 Nov 2024]

[20] Wang J,Xiao G,Li S.Data driven production modeling and simulation of complex automobile general assembly plant.Comput Ind2011;62:765-75

[21] Pozdnoukhov A,Kanevski M.Data-driven topo-climatic mapping with machine learning methods.Nat Hazards2009;50:497-518

[22] Srinivasan V.Low cycle fatigue and creep–fatigue interaction behavior of 316L(N) stainless steel and life prediction by artificial neural network approach.Int J Fatigue2003;25:1327-38

[23] Nasiri S,Weinberg K.Fracture mechanics and mechanical fault detection by artificial intelligence methods: a review.Eng Fail Anal2017;81:270-93

[24] Barbosa JF,Júnior RF.Fatigue life prediction of metallic materials considering mean stress effects by means of an artificial neural network.Int J Fatigue2020;135:105527

[25] Romano S,Gumpinger J,Beretta S.Qualification of AM parts: extreme value statistics applied to tomographic measurements.Mater Design2017;131:32-48

[26] Plessis A, Yadroitsava I, Yadroitsev I. Effects of defects on mechanical properties in metal additive manufacturing: a review focusing on X-ray tomography insights.Mater Design2020;187:108385

[27] Bhadeshia HKDH.Neural networks in materials science.ISIJ Int1999;39:966-79

[28] Zhou Y,Wang D.New trends in additive manufacturing of high-entropy alloys and alloy design by machine learning: from single-phase to multiphase systems.J Mater Inf2022;2:18

[29] Wang W,Chai Z.Big data-assisted digital twins for the smart design and manufacturing of advanced materials: from atoms to products.J Mater Inf2022;2:1

[30] Xu P,Li M.Virtual sample generation in machine learning assisted materials design and discovery.J Mater Inf2023;3:16

[31] Zhou Z,Yang Y.A critical review of the machine learning guided design of metallic glasses for superior glass-forming ability.J Mater Inf2022;2:2

[32] Hu Y,Withers P.The effect of manufacturing defects on the fatigue life of selective laser melted Ti-6Al-4V structures.Mater Design2020;192:108708

[33] Zhang M,Zhang X.High cycle fatigue life prediction of laser additive manufactured stainless steel: a machine learning approach.Int J Fatigue2019;128:105194

[34] Serrano-Munoz I,Mokso R,Nadot Y.Location, location &size: defects close to surfaces dominate fatigue crack initiation.Sci Rep2017;7:45239 PMCID:PMC5366922

[35] Benedetti M,Bandini M,Carmignato S.Low- and high-cycle fatigue resistance of Ti-6Al-4V ELI additively manufactured via selective laser melting: Mean stress and defect sensitivity.Int J Fatigue2018;107:96-109

[36] Ren S,Liu T.Effect of build orientation on mechanical properties and microstructure of Ti-6Al-4V manufactured by selective laser melting.Metall Mater Trans A2019;50:4388-409

[37] Barba D,Tang Y,Reed R.On the size and orientation effect in additive manufactured Ti-6Al-4V.Mater Design2020;186:108235

[38] Liu J,Liu J.Investigation of fatigue behavior of laser powder bed fusion Ti-6Al-4V: roles of heat treatment and microstructure.Int J Fatigue2023;176:107839

[39] Sharma S,Athaiya A.Activation functions in neural networks.Int J Eng Appl Sci Technol2020;4:310-6Available from: https://www.ijeast.com/papers/310. [Last accessed on 29 Nov 2024]

[40] Apicella A,Isgrò F.A survey on modern trainable activation functions.Neural Netw2021;138:14-32

[41] Rumelhart DE,Williams RJ.Learning representations by back-propagating errors.Nature1986;323:533-6

[42] Dieter GE. Mechanical metallurgy. 3rd edition. McGraw Hill, New York, 1986. Available from: https://books.google.com/books?id=4pZRAAAAMAAJ&hl=zh-CN&source=gbs_book_other_versions. [Last accessed on 29 Nov 2024]

[43] Murakami Y.Chapter 6 - Effects of nonmetallic inclusions on fatigue strength. In: Metal Fatigue. Elsevier; 2002. pp. 75-127.

[44] Stromeyer CE.The determination of fatigue limits under alternating stress conditions.Proc R Soc Lond A1914;90:411-25

[45] Paris P.A critical analysis of crack propagation laws.J Basic Eng1963;85:528-33

[46] Yukitaka M.Quantitative evaluation of fatigue strength of metals containing various small defects or cracks.Eng Fract Mech1983;17:1-15

[47] Deng K,Liu W,Zhao P.Probabilistic-based random maximum defect estimation and defect-related fatigue life prediction for laser direct deposited 316L parts.J Mater Process Tech2022;299:117389

[48] Liu F,Chen Y,Wang Q.Effects of defects on tensile and fatigue behaviors of selective laser melted titanium alloy in very high cycle regime.Int J Fatigue2020;140:105795

[49] Tang M.Fatigue life prediction for AlSi10Mg components produced by selective laser melting.Int J Fatigue2019;125:479-90

[50] Zhang M,Zhang X,Hardacre D.Predictive models for fatigue property of laser powder bed fusion stainless steel 316L.Mater Design2018;145:42-54

[51] Liu A,Liang Z.An evaluation on high cycle fatigue fracture characteristics of 2024-T351 Al alloy with different surface defects.Mech Mater2022;164:104133

[52] Chang D,Chen H,Wen J.Effects of defect size and location on high cycle fatigue life of a maraging stainless steel at ambient and cryogenic temperatures.Int J Fatigue2022;161:106906

[53] Tiryakioğlu M.Relationship between defect size and fatigue life distributions in Al-7 Pct Si-Mg alloy castings.Metall Mater Trans A2009;40:1623-30

[54] Queiroz H,Silva C.A coupled critical plane- area methodology to estimate fatigue life for an AISI 1045 steel with small artificial defects.Theor Appl Fract Mec2022;121:103426

[55] Murakami Y,Wada K.Essential structure of S-N curve: prediction of fatigue life and fatigue limit of defective materials and nature of scatter.Int J Fatigue2021;146:106138

[56] Li J,Qian G.Machine learning based very-high-cycle fatigue life prediction of Ti-6Al-4V alloy fabricated by selective laser melting.Int J Fatigue2022;158:106764

[57] Zhan Z.A novel approach based on the elastoplastic fatigue damage and machine learning models for life prediction of aerospace alloy parts fabricated by additive manufacturing.Int J Fatigue2021;145:106089

[58] Zhan Z.Machine learning based fatigue life prediction with effects of additive manufacturing process parameters for printed SS 316L.Int J Fatigue2021;142:105941

[59] Ayoub G,Zaïri F,Charrier P.Fatigue life prediction of rubber-like materials under multiaxial loading using a continuum damage mechanics approach: effects of two-blocks loading and R ratio.Mech Mater2012;52:87-102

[60] Darabi MK,Little DN.A continuum damage mechanics framework for modeling micro-damage healing.Int J Solids Struct2012;49:492-513