Prediction of residual stress in electron beam welding of stainless steel from process parameters and natural frequency of vibrations using machine-learning algorithms
Restricted accessResearch articleFirst published online June, 2021
Prediction of residual stress in electron beam welding of stainless steel from process parameters and natural frequency of vibrations using machine-learning algorithms
In the present study, machine learning algorithms have been used to predict residual stress during electron beam welding of stainless steel using the information of input process parameters and natural frequency of vibrations. Accelerating voltage, beam current and welding speed have been considered as input process parameters. Both residual stress and natural frequencies of vibration of the weld obtained using each set of the input parameters are measured experimentally. A number of machine learning algorithms, namely M5 algorithm-based Model Trees Regression, Random forest, Support Vector Regression, Reduced Error Pruning Tree, Multi-layer perceptron, Instance-based k-nearest neighbor algorithm, and Locally weighted learning have been used for the said purpose. Support vector regression and Locally weighted learning are found to perform consistently good and bad, respectively. The predicted welding residual stresses have been validated experimentally through X-ray diffraction (XRD) and good agreements are obtained. In addition, statistical tests are conducted, and the estimated reliability values of the employed models are analyzed through Monte-Carlo simulations.
VieiraABRadeDAScottiA.Identification of welding residual stresses in rectangular plates using vibration responses. Inverse Probl Sci Eng2006;
14: 313–331.
2.
MathewJGriffinJAlamaniotisM, et al.
Prediction of welding residual stresses using machine learning: comparison between neural networks and neuro-fuzzy systems. Appl Soft Comput J2018;
70: 131–146.
3.
DasDPratiharDKRoyGG, et al.
Phenomenological model-based study on electron beam welding process, and input-output modeling using neural networks trained by back-propagation algorithm, genetic algorithms, particle swarm optimization algorithm and bat algorithm. Appl Intell2018;
48: 2698–2718.
4.
RaiRPalmerTAElmerJW, et al.
Heat transfer and fluid flow during electron beam welding of 304L stainless steel alloy. Weld J2009;
88: 54–61.
5.
HouWWeiYJinY, et al.
Deep features based on a DCNN model for classifying imbalanced weld flaw types. Measurement2019;
131: 482–489.
6.
VournaPKtenaATsakiridisPE, et al.
An accurate evaluation of the residual stress of welded electrical steels with magnetic Barkhausen noise. Measurement2015;
71: 31–45.
7.
LuFLiXLiZ, et al.
Formation and influence mechanism of keyhole-induced porosity in deep-penetration laser welding based on 3D transient modeling. Int J Heat Mass Transf2015;
90: 1143–1152.
8.
Akbari MousaviSAAMiresmaeiliR.Experimental and numerical analyses of residual stress distributions in TIG welding process for 304L stainless steel. J Mater Process Technol2008;
208: 383–394.
9.
ParadowskaAPriceJWHIbrahimR, et al.
A neutron diffraction study of residual stress due to welding. J Mater Process Technol2005;
164–165: 1099–1105.
10.
ShiJBoothS.An investigation of effects of welding residual stresses on creep crack growth for a low alloy butt weld. Int J Press Vessel Pip2013;
108–109: 67–71.
11.
LiuCZhangJXXueCB.Numerical investigation on residual stress distribution and evolution during multipass narrow gap welding of thick-walled stainless steel pipes. Fusion Eng Des2011;
86: 288–295.
12.
RaoSS.Mechanical vibrations in SI units. 6th ed.
London:
Pearson Education, Inc., 2017.
13.
BezerraACVieiraLCRadeDA, et al.
On the influence of welding residual stresses on the dynamic behavior of structures. Shock Vib2008;
15: 447–458.
JoyVVJCIVargheseVM.Effect of welding residual stress on natural frequency of structures. Int J Innov Res Adv Eng2014;
1: 459–462.
16.
BadgujarBPKumarSJhaMN, et al.
An investigation of electron beam welding of Nb-1Zr-0 . 1C alloy : process parameters and microstructural analysis. J Manuf Process2017;
28: 326–335.
17.
JubbJEMPhillipsIGBeckerH.Interrelation of structural stability, stiffness, residual stress and natural frequency. J Sound Vib1975;
39: 121–134.
18.
KaldasMMDickinsonSM.Vibration and buckling calculations for rectangular plates subject to complicated in-plane stress distributions by using numerical integration in a Rayleigh-Ritz analysis. J Sound Vib1981;
75: 151–162.
19.
DasDPratiharDKRoyGG. Establishing a correlation between residual stress and natural frequency of vibration for electron beam ButtWeld of AISI 304 stainless steel. Arab J Sci Eng. Epub ahead of print 2020. DOI: 10.1007/s13369-020-04560-0.
20.
Lostado LR, Corral BM, Martínez CM, et al. Residual stresses with time-independent cyclic plasticity in finite element analysis of welded joints. Metals 2017; 7(4): 136.
21.
OlabiAGLostadoRBenyounisKY. Review of microstructures, mechanical properties, and residual stresses of ferritic and martensitic stainless-steel welded joints. Epub ahead of print 2014. DOI: 10.1016/B978-0-08-096532-1.00617-8.
22.
GiriAPandeyCMahapatraMM, et al.
On the estimation of error in measuring the residual stress by strain gauge rosette. Meas J Int Meas Confed2015;
65: 41–49.
23.
PandeyCMahapatraMMKumarP, et al.
Effect of weld consumable conditioning on the diffusible hydrogen and subsequent residual stress and flexural strength of multipass welded P91 steels. Metall Mater Trans B Process Metall Mater Process Sci2018;
49: 2881–2895.
24.
SatpathyMPMishraSBSahooSK.Ultrasonic spot welding of aluminum-copper dissimilar metals: a study on joint strength by experimentation and machine learning techniques. J Manuf Process2018;
33: 96–110.
25.
LostadoRMartinezRFMac DonaldBJ, et al.
Combining soft computing techniques and the finite element method to design and optimize complex welded products. Integr Comput Aided Eng2015;
22: 153–170.
26.
EscribanoRLostadoRMartínez-de-PisónFJ, et al.
Modelling a skin-pass rolling process by means of data mining techniques and finite element method. J Iron Steel Res Int2012;
19: 43–49.
27.
LostadoREscribano GarcíaRFernandez MartinezR.Optimization of operating conditions for a double-row tapered roller bearing. Int J Mech Mater Des2016;
12: 353–373.
28.
DasDPalARDasAK, et al.
Nature-inspired optimization algorithm-tuned feed-forward and recurrent neural networks using CFD-based phenomenological model-generated data to model the EBW process. Arab J Sci Eng2020;
45: 2779–2797.
29.
Fernandez MartinezRLostado LorzaRSantos DelgadoAA, et al.
Optimizing presetting attributes by softcomputing techniques to improve tapered roller bearings working conditions. Adv Eng Softw2018;
123: 13–24.
30.
GómezFSLorzaRLBobadillaMC, et al.
Improving the process of adjusting the parameters of finite element models of healthy human intervertebral discs by the multi-response surface method. Materials (Basel)2017;
10: 1–39.
31.
LostadoRMartínez-De-PisónFJFernándezR, et al.
Using genetic algorithms to optimize the material behaviour model in finite element models of processes with cyclic loads. J Strain Anal Eng Des2011;
46: 143–159.
32.
FengSZhouHDongH.Using deep neural network with small dataset to predict material defects. Mater Des2019;
162: 300–310.
33.
VermaSGuptaMMisraJP.Performance evaluation of friction stir welding using machine learning approaches.MethodsX2018;
5: 1048–1058.
34.
ChenJWangTGaoX, et al.
Real-time monitoring of high-power disk laser welding based on support vector machine. Comput Ind2018;
94: 75–81.
35.
MunirNKimHParkJ, et al. Convolutional neural network for ultrasonic weldment flaw classification in noisy conditions. Ultrasonics. Epub ahead of print 2018. DOI: 10.1016/j.ultras.2018.12.001.
36.
PetkoviD.Prediction of laser welding quality by computational intelligence approaches. Optik (Stuttg)2017;
140: 597–600.
37.
LiaoTWTriantaphyllouEChangPC.Detection of welding flaws with MLP neural network and case based reasoning. Intell Autom Soft Comput2003;
4: 259–267.
38.
GhantyPPaulSMukherjeeDP, et al.
Modelling weld bead geometry using neural networks for GTAW of austenitic stainless steel. Sci Technol Weld Join2007;
12: 649–658.
39.
WangGLiaoTW.Automatic identification of different types of welding defects in radiographic images. Ndt E Int2002;
35: 519–528.
40.
LiangRYuRLuoY, et al.
Machine learning of weld joint penetration from weld pool surface using support vector regression. J Manuf Process2019;
41: 23–28.
41.
BoerschIFüsselUGreschC, et al.
Data mining in resistance spot welding: a non-destructive method to predict the welding spot diameter by monitoring process parameters. Int J Adv Manuf Technol2018;
99: 1085–1099.
42.
DasDPratiharDKRoyGG.Cooling rate predictions and its correlation with grain characteristics during electron beam welding of stainless steel. Int J Adv Manuf Technol2018;
97: 2241–2254.
43.
DasDPratiharDKRoyGG.Effects of space charge on weld geometry and cooling rate during electron beam welding of stainless steel. Optik (Stuttg)2020;
206: 163722.
44.
DasDPratiharDKRoy GG.Electron beam melting of steel plates: temperature measurement using thermocouples and prediction through finite element analysis. In: MandalDKSyanCS (eds) CAD/CAM, robotics and factories of the future.
New Delhi:
Springer, 2016, pp. 579–588.
45.
GamagePJayamahaNPGriggNP.Acceptance of Taguchi’s quality philosophy and practice by lean practitioners in apparel manufacturing. Total Qual Manag Bus Excell2017;
28: 1322–1338.
46.
FerroPZambonABonolloF.Investigation of electron-beam welding in wrought Inconel 706 – experimental and numerical analysis. Mater Sci Eng A2005;
392: 94–105.
47.
NatarajJRRajkumarGR.Finite element analysis of dissimilar welding between newly developed Cr-free nickel based welding electrode and stainless steel AISI 304. Int J Res Eng Technol2013;
2: 130–135.
48.
GrayTCamilleriDMcPhersonN.Control of welding distortion in thin-plate fabrication. 1st ed.
Cambridge:
Woodhead Publishing Limited, 2014.
49.
CullityBDStockSR.Elements of X-ray diffraction. 3rd ed.London:
Pearson Education India, 2014.
50.
ÖzçiftA.Forward stage-wise ensemble regression algorithm to improve base regressors prediction ability: an empirical study. Expert Syst2014;
31: 1–8.
51.
WittenIHFrankEHallMA, et al. Data mining: Practical machine learning tools and techniques. 2nd ed.
San Francisco:
Morgan Kaufmann, 2005.
52.
GraczykMLasotaTTrawinskiB.Comparative analysis of premises valuation models using KEEL, RapidMiner, and WEKA. In: Nguyen NT, Kowalczyk R and Chen SM (eds) International conference on computational collective intelligence.
Berlin:
Springer, 2009, pp. 800–812.
53.
ShanmuganathanSNarayananAMohamedM, et al. Recent advances on soft computing and data Mining. Johor, Malaysia. Epub ahead of print 2014. DOI: 10.1007/978-3-319-07692-8.
54.
HsiehCHLuRHLeeNH, et al.
Novel solutions for an old disease: Diagnosis of acute appendicitis with random forest, support vector machines, and artificial neural networks.Surgery2011;
149: 87–93.
55.
AmrehnMMuallaFAngelopoulouE, et al. The random forest classifier in WEKA: discussion and new developments for imbalanced data. arXiv Prepr arXiv181208102, http://arxiv.org/abs/1812.08102 (2018).
56.
NamlıEErdalHErdalHI.Artificial intelligence-based prediction models for energy performance of residential buildings. In: BalkayaNGuneysuS (eds) Recycling and reuse approaches for better sustainability.
Cham:
Springer International Publishing, p. 304.
57.
VapnikV.The nature of statistical learning theory.
New York:
Springer-Verlag Inc, 2013.
58.
ZhanCGanAHadiM.Prediction of lane clearance time of freeway incidents using the M5P tree algorithm. IEEE Trans Intell Transp Syst2011;
12: 1549–1557.
59.
AlamIFaridDMRossettiRJF.The prediction of traffic flow with regression analysis. In: Abraham A, Dutta P, Mandal J, Bhattacharya A and Dutta S (eds) Emerging technologies in data mining and information security.
Singapore:
Springer, pp. 661–671.
60.
LostadoRMartínez-De-PisónFJPerníaA, et al.
Combining regression trees and the finite element method to define stress models of highly non-linear mechanical systems. J Strain Anal Eng Des2009;
44: 491–502.
61.
MartínÓPeredaMSantosJI, et al.
Assessment of resistance spot welding quality based on ultrasonic testing and tree-based techniques. J Mater Process Technol2014;
214: 2478–2487.
62.
XingBXiaoYQinQH, et al.
Quality assessment of resistance spot welding process based on dynamic resistance signal and random forest based. Int J Adv Manuf Technol2018;
94: 327–339.
63.
GöndörMBresfeleanVP.REPTree and M5P for measuring fiscal policy influences on the Romanian capital market during 2003–2010. Int J Math Comput Simul2012;
6: 378–386.
64.
ShafiullahGMThompsonAWolfsPJ, et al. Reduction of power consumption in sensor network applications using machine learning techniques. In: IEEE Reg 10 Annu Int Conf Proceedings/TENCON. Epub ahead of print 2008. DOI: 10.1109/TENCON.2008.4766574.
65.
ZaharaSIIlyasMZiaT.A study of comparative analysis of regression algorithms for reusability evaluation of object oriented based software components. ICOSST 2013–2013 Int Conf Open Source Syst Technol Proc, Lahore, Pakistan, 16–18 December 2013, pp.75–80. IEEE.
66.
SchaalSAtkesonCG.Scalable techniques from nonparametric statistics. Appl Intell2002;
16: 49–60.
67.
Schaal S, Atkeson CG, Vijayakumar S. Real-time robot learning with locally weighted statistical learning. Proceedings 2000 ICRA. Millennium Conference. IEEE International Conference on Robotics and Automation. Symposia Proceedings (Cat. No.00CH37065), San Francisco, CA, USA, 24–28 April 2000, pp. 288–293. IEEE.
68.
MajiK.Analysis and synthesis of sheet metal forming by laser heating using finite element method and soft computing. Doctoral dissertation, IIT Kharagpur, http://www.idr.iitkgp.ac.in/xmlui/handle/123456789/3780 (2014, accessed 31 July 2020).
69.
Haykin S. Neural networks- a comprehensive foundation. Vol. 2. Upper Saddle River, NJ, USA: Prentice-Hall, 2004.
70.
NaMGKimJWLimDH, et al.
Residual stress prediction of dissimilar metals welding at NPPs using support vector regression. Nucl Eng Des2008;
238: 1503–1510.
71.
FarfaniHABehnamfarFFathollahiA.Dynamic analysis of soil-structure interaction using the neural networks and the support vector machines. Expert Syst Appl2015;
42: 8971–8981.
72.
ThankachanTSooryaprakashK.Artificial neural network-based modeling for impact energy of cast duplex stainless steel. Arab J Sci Eng2018;
43: 1335–1343.
MorleyKCGallateJEHuntGE, et al.
Increased anxiety and impaired memory in rats 3 months after administration of 3,4-methylenedioxymethamphetamine (‘Ecstasy’). Eur J Pharmacol2001;
433: 91–99.
75.
HayterAJ.A proof of the conjecture that the Tukey-Kramer multiple comparisons procedure is conservative. Ann Stat1984;
12: 61–75.
76.
DerracJGarcíaSMolinaD, et al.
A practical tutorial on the use of nonparametric statistical tests as a methodology for comparing evolutionary and swarm intelligence algorithms. Swarm Evol Comput2011;
1: 3–18.
77.
TakeshiM.The Monte-Carlo simulation method for system reliability and risk analysis. Nucl Saf Simul2013;
4: 44–52.
