مهندسی متالورژی

مهندسی متالورژی

مقایسه قابلیت مدلهای آرنیوسی و شبکه مصنوعی عصبی در مدلسازی رفتار سیلان سوپرالیاژ اینکونل 625 تولید شده به روش ساخت افزایشی

نوع مقاله : مقاله پژوهشی

نویسندگان
حمیدرضا عابدی
سبحان نجفی
دانشکده مهندسی مواد و متالورژی، دانشگاه علم و صنعت ایران، تهران، ایران
10.22076/me.2024.1998893.1408
چکیده
ساخت محصولات نزدیک به شکل نهایی از جنس آلیاژهای پیشرفته پراستحکام از طریق روشهای ساخت افزایشی، توجه بسیاری از صنعتگران و محققان را به خود جلب کرده است. آلیاژهای پیشرفته مهندسی مانند سوپرآلیاژهای پایه نیکل جهت سرویس دهی در شرایط خاص مانند دمای بالا، زمانهای طولانی مدت، و شرایط خورنده و اکسید شونده طراحی می شوند. بنابراین ضرورت بررسی خواص مکانیکی دمای بالای آلیاژهای پیشرفته پراستحکام ساخته شده به روش ساخت افزایشی کاملا محسوس است. در این پژوهش رفتار سیلان دمای بالای سوپر آلیاژ Inconel625 تولید شده به روش ذوب لیزری انتخابی مورد بررسی قرار گرفته است. به این منظور آزمایش‌های فشار در دماهای 800 ، 900 ، 1000 و ℃1100 تحت نرخ کرنشهای 001/0، 01/0، و s-11/0 انجام شد. جهت مدلسازی رفتارسیلان آلیاژ از مدل‌ رفتاری آرنیوسی اصلاح شده و مدل شبکه عصبی مصنوعی استفاده شد. برای ارزیابی دقت مدلهای به‌کارگرفته‌شده از پارامترهای استاندارد آماری در قالب ضریب همبستگی، ریشه میانگین مربعات خطا، و میانگین خطای نسبی مطلق استفاده شد. با توجه به تغییرات قابل توجه ضریب حساسیت به نرخ کرنش و انرژی فعالسازی تغییرشکل در محدوده دمایی مورد بررسی، مدل آرنیوسی نتوانست پیش بینی دقیقی از رفتار سیلان داشته باشد. در حالی که مدل شبکه عصبی مصنوعی از توانایی بسیار بالایی در مدلسازی رفتار سخت‌شوندگی و نرم‌شوندگی سیلان سوپرآلیاژ مورد بررسی برخوردار است. بر اساس داده های پیش بینی شده نقشه‌های بازده و اتلاف توان در کرنشهای 1/0 ، 2/0 ، 3/0 و 4/0 رسم شدند و مناطق ناپایداری متالورژیکی و شرایط بهینه تغییرشکل مورد شناسایی قرار گرفت.
کلیدواژه‌ها
سوپرآلیاژ پایه نیکل
ساخت افزایشی
رفتار سیلان
نقشه فرایند
شبکه عصبی مصنوعی

عنوان مقاله English

Comparing the capability of Arrhenius type and artificial neural network models in predicting the high temperature flow behavior of selective laser melted Inconel 625 alloy

نویسندگان English

Hamid Reza Abedi
Sobhan Asl Najafi
School of Metallurgy & Materials Engineering, Iran University of Science and Technology (IUST), Tehran, Iran
چکیده English

Conventional methods face limitations when it comes to producing near-net-shape products of high-strength advanced engineering alloys, while additive manufacturing as an alternative approach have garnered the interest of artisans and researchers. High-performance engineering materials, such as nickel-based superalloys, are specifically formulated to endure harsh conditions like elevated temperatures, extended exposure time, and corrosive surroundings. Consequently, there is a significant demand to study the high temperatures mechanical characteristics of advanced high-strength alloys produced through additive manufacturing. In this research, the high temperature flow behavior of Inconel625 superalloy produced through selective laser melting has been investigated. For this purpose, hot compression tests were performed at temperatures of 800, 900, 1000, and 1100 ℃ under the strain rates of 0.001, 0.01, and 0.1s-1. In order to modeling the flow stress behavior, modified Arrhenius-type model and artificial neural network model were employed. Standard statistical parameters in the form of correlation coefficient (R), root mean square error (RMSE) and average absolute relative error (AARE) were used to evaluate the accuracy of the developed models. Due to the significant changes in the strain rate sensitivity and deformation activation energy in the studied temperature range, Arrhenius model could not accurately predict the high temperature flow behavior. While the artificial neural network model represented high capability in modeling the hardening and softening behavior of the investigated superalloy. Based on the predicted data, power efficiency and dissipation maps were constructed at strains of 0.1, 0.2, 0.3, and 0.4, then the metallurgical instability regions and optimal deformation conditions were identified.

کلیدواژه‌ها English

Nickel-based superalloy
Additive Manufacturing
Flow behavior
Processing map
Artificial neural network
[1] Kale AB, Alluri P, Singh AK, Choi SH, The deformation and fracture behavior of 316L SS fabricated by SLM under mini V-bending test, Int. J. Mech. Sci. 2021, 196: 106292, https://doi.org/10.1016/j.ijmecsci.2021.106292

[2] Roshani M, Abedi HR, Saboori A, Comparing the Cold, Warm, and Hot Deformation Flow Behavior of Selective Laser-Melted and Electron-Beam-Melted Ti–6Al–2Sn–4Zr–2Mo Alloy, Adv. Eng. Mater. 2024, 26: 2301046, https://doi.org/10.1002/adem.202301046

[3] Abbasi Z, Abedi HR, Dynamic recrystallization behavior of selective laser melted 316L stainless steel, J. Mater. Res. Tech. 2023, 27: 6458-6473, https://doi.org/10.1016/j.jmrt.2023.11.120

[4] Tabatabaei N, Zarei-Hanzaki A, Moshiri A, Abedi HR, The effect of heat treatment on the room and high temperature mechanical properties of AlSi10Mg alloy fabricated by selective laser melting, J. Mater. Res. Tech. 2023, 23: 6039-6053, https://doi.org/10.1016/j.jmrt.2023.02.086

[5] Habibiyan A, Hanzaki A, Abedi HR, An investigation into microstructure and high-temperature mechanical properties of selective laser-melted 316L stainless steel toward the development of hybrid Ampliforge process, Int. J. Adv. Manuf. 2020, 110: 383-394, https://doi.org/10.1007/s00170-020-05870-1

[6] Rozegari Ghashghay B, Abedi HR, Shabestari SG, On the capability of grain refinement during selective laser melting of AlSi10Mg alloy, J. Mater. Res. Tech. 2023, 24: 9722-9730, https://doi.org/10.1016/j.jmrt.2023.05.122

[7] Khodashenas H, Mirzadeh H, Post-processing of additively manufactured high-entropy alloys - A review, J. Mater. Res. Tech. 2022, 21: 3795-3814, https://doi.org/10.1016/j.jmrt.2022.11.027

[8] Su R, Hao D, He P, Wu D, Wang Q, Dong H, Ma H, Effect of Co on creep and stress rupture properties of nickel-based superalloys – A review, J. All. Comp. 2023, 967: 171744

https://doi.org/10.1016/j.jallcom.2023.171744

[9] Ghaemifar S, Mirzadeh H, Precipitation kinetics of niobium carbide (NbC) during homogenization heat treatment of additively manufactured inconel 718 superalloy, J. Mater. Res. Tech. 2023, 25: 1774-1781, https://doi.org/10.1016/j.jmrt.2023.06.069

[10] Ibrahim M, Saeed T, Alshehri AM, Chu YM, Using artificial neural networks to predict the rheological behavior of non-Newtonian graphene–ethylene glycol nanofluid, J. Therm. Anal. Calorim. 2021, 145: 1925-1934, https://doi.org/10.1007/s10973-021-10682-w

[11] Rahimi I, Abedi HR, Predicting the strain hardening behavior and constructing processing maps of a lightweight Fe–Mn–Al–C steel through phenomenological modelling, J. Mater. Res. Tech. 2024, 30: 2891-2901, https://doi.org/10.1016/j.jmrt.2024.04.002

[12] Boluri Gashti AH, Abedi HR, Salehi MT, Microstructure evolution and constitutive modeling of as-cast A356 aluminum alloy in semi-solid deformation regime, J. Mater. Res. Tech. 2023, 24: 7720-7731, https://doi.org/10.1016/j.jmrt.2023.05.025

[13] Feng W, Qin F, Long H, Hot workability analysis and processing parameters optimisation for 20CrMnTiH steel by combining processing map with microstructure, Ironmak. Steelmak. 2018, 45: 317-324, https://doi.org/10.1080/03019233.2016.1264145

[14] Esmaeilpour A, Abedi HR, Comparing the Capability of Phenomenological (Johnson-Cook and Arrhenius-Type) and Artificial Neural Network Models in Predicting the Hot Deformation Behavior of Additively Manufactured 316L Stainless Steel, Iranian J. of Mater. Form. 2023, 9: 62-73, https://doi.org/10.22099/ijmf.2022.43705.1231

[15] Sohrabi MJ, Mirzadeh H, Sadeghpour S, Mahmudi R, Johnson–Mehl–Avrami–Kolmogorov (JMAK-type) Analysis of Deformation-Induced Martensitic Transformation in AISI 304L Stainless Steel, Iranian J. Mater. Sci. and Eng. 2023, 20:1-9, , https://doi.org/10.22068/ijmse.3335

[16] Zamani MR, Mirzadeh H, Malekan M, Artificial neural network applicability in studying hot deformation behaviour of high-entropy alloys, Mater Sci and Tech 2023, 39: 3351-3359, https://doi.org/10.1080/02670836.2023.2231767
[17] M. Karimzadeh a, M. Malekan a, H. Mirzadeh a, N. Saini b, L. Li, Hot deformation behavior analysis of as-cast CoCrFeNi high entropy alloy using Arrhenius-type and artificial neural network models, Intermetallics. 2024, 168: 108240, https://doi.org/10.1016/j.intermet.2024.108240
[18] Esmaeilpour A, Abedi HR, Mirzaei A, Habibiyan A, Constructing the high temperature efficiency and instability maps of selective laser melted 316L stainless steel through artificial neural network modeling, J. Mater. Res. Tech. 2022, 18: 4578-4589, https://doi.org/10.1016/j.jmrt.2022.04.116
[19] Zamani MR, Roostaei M, Mirzadeh H, Malekan M, Song M, Tailoring the microstructure and mechanical properties of (CrMnFeCoNi)100-xCx high-entropy alloys: Machine learning, experimental validation, and mathematical modeling, Curr. Opin. Solid State Mater. Sci. 2023, 27: 101105, https://doi.org/10.1016/j.cossms.2023.101105
[20] He DG, Lin YC, Chen MS, Chen J, Wen DX, Chen XM, Effect of pre-treatment on hot deformation behavior and processing map of an aged nickel-based superalloy, J. All. Comp. 2015, 649: 1075-1084, https://doi.org/10.1016/j.jallcom.2015.07.213
[21] Wen DX, Lin YC, Li HB, Chen XM, Deng J, Li LT, Hot deformation behavior and processing map of a typical Ni-based superalloy, Mater. Sci. Eng. A 2014, 591: 183-192, https://doi.org/10.1016/j.msea.2013.09.049
[22] Huang A, Wang Z, Yuan Q, Chen R, Zhang Y, Guan R, Study on serration flow and dynamic strain aging of Cr–Ti–B low-carbon steel, J. Mater. Res. Tech. 2021, 12: 1543-1551, https://doi.org/10.1016/j.jmrt.2021.03.100
[23] Moris Devotta A, Sivaprasad, PV, Beno T, Eynian M, Hjertig K, Magnevall M, Lundblad M, A Modified Johnson-Cook Model for Ferritic-Pearlitic Steel in Dynamic Strain Aging Regime, J. Met. 2019, 9: 528, https://doi.org/10.3390/met9050528
[24] Rezaei Ashtiani HR, Shahsavari P, A comparative study on the phenomenological and artificial neural network models to predict hot deformation behavior of AlCuMgPb alloy, J. Alloys Compd. 2016, 687: 263-273, https://doi.org/10.1016/j.jallcom.2016.04.300
[25] Sabokpa O, Zarei-Hanzaki A, Abedi HR, and Haghdadi N, Artificial neural network modeling to predict the high temperature flow behavior of an AZ81 magnesium alloy, Mater. Des. 2012, 39: 390-396, https://doi.org/10.1016/j.matdes.2012.03.002
[26] Chen L, Luo R, Yang Y, Peng CT, Gui X, Zhang J, Song K, Gao P & Cheng X , Investigation on the Hot Deformation Behavior of 316L Stainless Steel Using 3D Processing Map, T Indian I Metals 2019, 72: 2997-3006, https://doi.org/10.1007/s12666-019-01674-4
[27] Babo NS, Tiwari SB, Rao B, Modified instability condition for identification of unstable metal flow regions in processing maps of magnesium alloys, Mater. Sci. and Tech. 2005, 21: 976-984, https://doi.org/10.1179/174328405X47519

  • تاریخ دریافت 26 تیر 1403
  • تاریخ بازنگری 29 مرداد 1403
  • تاریخ پذیرش 03 شهریور 1403