Measurement of delta ferrite in cladding gives important insight into the future mechanical and corrosion resistant behaviour of the cladded structures. The amount of delta ferrite formed during cladding is influenced by process parameters such as welding speed, welding current, and nozzle-to-plate distance. Therefore, it is essential to predict the effect of these parameters on the formation of delta ferrite. This article discusses the development of an artificial neural network model to predict the delta ferrite content in austenitic stainless-steel claddings deposited by the flux cored arc welding process. A novel approach of using the design of experiments to collect data to train the network has been adopted in this investigation. The study revealed that the delta ferrite content can be predicted more accurately using the neural networks with a minimum number of experiments. The results also indicated that welding current and speed have a significant influence on the amount of ferrite and the interaction effects of these parameters play a major role in determining ferrite in the claddings.
MotarjemiA. K.KocakM., Fracture assessment of weld repaired cladded steel wide plates by SINTAP defect assessment procedureWeld. World45 (2001): 27–33.
2.
MuruganN.ParmerR. S., Mathematical models for bead geometry prediction in austenitic stainless steel surfacing by MIG weldingInt. J. Joining Mater.72/3 (1995): 71–80.
3.
DavisJ. R., ASM specialty handbook on stainless steel (New York, USA: ASM International1996) 2nd print.
4.
AlamN.JarvisL.HarrisD.SoltaA., Laser cladding for repair of engineering componentsAust. Weld. J.472 (2002): 38–47.
5.
HestonT., Cladding operations doubles life of boiler tubingWeld. J.79 (2000): 45–47.
6.
PalaniP. K.MuruganN., Development of mathematical model for prediction of weld bead geometry in cladding by flux cored arc weldingInt. J. Adv. Manuf. Technol.307—8 (2006): 669–676.
7.
CuiY.LundinC. D., Ferrite number as a function of the larson—miller parameter for austenitic stainless weld metals after creep testingMetall. Mater. Trans. A35A (2004): 3631–3634.
8.
LeeJ. S.KimI.KimuraA., Application of small punch test to evaluate sigma-phase embitterment of pressure vessel cladding materialJ. Nucl. Sci. Technol.40 (2003): 664–671.
9.
DhanukaM. P., Measurement of delta ferrite content in austenitic stainless steel weld metals using WRC-92 diagram (2003).
10.
OlsonD. L., Prediction of austenitic weld metal microstructure and propertiesWeld. J.64 (1985): 281s–295s.
11.
SmithD. V., A practical approach to ferrite in stainless steel weld metalWeld. J.67 (1988): 57–69.
12.
BuvanasekaramG.PrasadV. N., Effect of cladding parameters on intergranular corrosion and bonding properties of cladding stainless steel layers proceedings of national conference on recent advances in materials processing (India: Annamalai University2001): 136–147.
13.
KaneS. F.FarlandA. L.SiewertT. A.McCowanC. N., Welding consumable development for a cryogenic (4 K) applicationWeld. J.78 (1999): 292s–300s.
14.
SaitoM.HoriyaT.YamamotoA.IidaT.OgataT., The effect of delta-ferrite phase in weld metal of stainless steel on the mechanical properties at cryogenic temperature Materials and welding research lab, Nagasaki-R7d Centre, available from Japan. http://www.enaa.or.jp/we-net/ronbun/1996/e16/naga1996.html.
VitekJ. M.IskanderY. S.OblowE. M., Improved ferrite number prediction in stainless steel arc welds using artificial neural networks — part 1: Neural network developmentWeld. J.79 (2000): 34s–40s.
17.
VitekJ. M.IskanderY. S.OblowE. M., Improved ferrite number prediction in stainless steel arc welds using artificial neural networks — part 2: Neural network developmentWeld. J.79 (2000): 41s–50s.
18.
KoteckiD. J., Standards and industrial methods for ferrite measurementWeld. J.77 (1998): 49–52.
19.
RaoP. K.KumarP. S., Assessment criterion for variability of delta ferrite in austenitic weld and clad metalsWeld. J.63 (1984): 231s–239s.
20.
NishimotoK.OgawaK., Corrosion properties in weldments of stainless steels (1) metallurgical factors affecting corrosion propertiesWeld. J.78 (1999): 2–11.
21.
KoteckiD. J., Ferrite determination in stainless steel welds — advances since 1974Weld. J.76 (1997): 24s–37s.
22.
SiewertT. A.McCowanC. N.OlsonD. L., Ferrite number prediction to 100 FN in stainless steel weld metalWeld. J.67 (1988): 289s–296s.
23.
BalmforthM. C.LippoldJ. C., A new ferritic-martensitic stainless steel constitution diagramWeld. J.79 (2000): 339s–345s.
24.
RaoP. K.KrishnanN.RaoR. V.RajaGuru G. J., Effect of welding parameters on the content of delta ferrite in austenitic stainless steel weld and clad metals (1986).
25.
VitekJ. M.DavidS. A.HinmanC. R., Improved ferrite number prediction model that accounts for cooling rate effects — part 1: Model developmentWeld. J.82 (2003): 10s–17s.
26.
VitekJ. M.DavidS. A.HinmanC. R., Improved ferrite number prediction model that accounts for cooling rate effects — part 2: Model resultsWeld. J.82 (2003): 43s–50s.
27.
LippoM. I.HazarabedianA.GarcOvejero J., Effects of delta ferrite on hydrogen embrittlement of austenitic stainless steel weldsCorros. Sci.41 (1999): 87–103.
28.
PerezT. E.FunesA., J. Ovejero Garcia, (1982).
29.
MudaliKamachi U.DayalR. K., Pitting corrosion resistance of as welded and thermally aged nitrogen containing type 316 stainless steel weld metalMater. Sci. Technol.16 (2000): 393–398.
30.
SunZ.KuoM.AnneergenI. Y.PanD., Effect of dual torch techniques on duplex stainless steel weldsInt. J. Mater. Sci. Eng.A356 (2003): 274–282.
31.
MuruganN.ParmerR. S., Effect of submerged arc process variables on dilution and bead geometry in single wire surfacingJ. Mater. Process. Technol.37 (1995): 767–780.
32.
MuruganN.ParmerR. S., Stainless steel cladding deposited by automatic gas metal arc weldingWeld. J.76 (1997): 391s–402s.
33.
MuruganN.PalaniP. K., Optimization of bead geometry in automatic stainless steel cladding by MIG welding using a genetic algorithmJ. Inst. Eng.84 (2004): 49–54 (India).
34.
KimI. S.SonK. J.YangY. S.YaragadaP. K. D. V., Sensitivity analysis for process parameters in GMA welding process using factorial design methodInt. J. Mach. Tools Manuf.43 (2003): 763–769.
35.
MontgomeryD. C., Design and analysis of experiments (New York: John Wiley & Sons Inc.2001).
36.
RmasamyS.GouldJ.WorkmanD., Design of experiments study to examine the effect of polarity on stud weldingWeld. J.81 (2002): 19s–26s.
37.
ChenS. B.LouY. J.WuL.ZhaoD. B., Intelligent methodology for sensing, modelling and control of pulsed GTAW: Part I — bead-on-plate weldingWeld. J.79 (2000): 151s–163s.
38.
PalaniP. K.MuruganN., Sensitivity of delta ferrite formation to FCAW process parameters in stainless steel claddingsPaton Weld. J.3 (2006): 16–21.
39.
IrvingB., Neural networks are paying off on the production lineWeld. J.76 (1997): 59–63.
40.
RidingsG. E.ThomsonR. C.ThewlisG., Prediction of multiwire submerged arc weld bead shape using neural network modelingSci. Technol. Weld. Joining7 (2002): 265–276.
41.
AndersonK.CookG. E.KarsaiG.RamasamyK., Artificial neural networks applied for welding process modeling and controlIEEE Trans. Ind. Appl.26 (1990): 824–830.
GaoJ.WuC., Neurofuzzy control of weld penetration in gas tungsten arc weldingSci. Technol. Weld. Joining82 (2003): 143–148.
44.
SaiKumarS.DhasharathR.KaulI. K.MohanramP. V., Development of model for high speed milling using neural networksManuf. Technol. Today Nov (2003): 3–7.
45.
AmesN.RambergM.JohnsonM.JohnsT., Comparison of austenitic, super austenitic and super duplex weld properties produced using GTAW flux (2001).
46.
KoteckiD. J., A martensite boundary on the WRC-1992 diagramWeld. J.78 (1999): 180s–192s.
47.
BhuvanasekaranG.PrasadV. N., Influence of process parameters on the quality of different types of stainless steel strip cladded metalWRI J.18 (1997): 136–158.
48.
GunasekaranB.VenkatG.DhandapaniK. P., Application of submerged arc strip cladding technology for surfacingWRI J.191 (1998): 5–9.
49.
KoteckiD. J.SiewertT. A., WRC-1992 constitution diagram for stainless steel weld metals: A modification of the WRC-1988 diagramWeld. J.71 (1992): 171s–178s.
50.
KangM. J.KimY.AhnS.RheeS., Spatter rate estimation in the short circuit transfer region of GMAWWeld. J.82 (2003): 238s–247s.
51.
WangF.LiZ., Predicting the composition of flux-cored wire cladded metal by a neural networkChina Weld.10 (2001): 44–50.
52.
DasS.BandyopadhyayP. P.ChattopadhyayA. B., Neural networks based tool wear monitoring in turning medium carbon steel using a coated carbide toolJ Mater. Process. Technol.63 (1997): 187–192.
NgS.-C.CheungC.-C.LeungS.-H., Magnified gradient function with deterministic weight modification in adaptive learningIEEE Trans. Neural Netw.15 (2004): 1411–1411.
55.
RumelhartD. E.HintonG. E.WilliamsR. J., Learning internal representations by error propagationParallel distributed processing (Cambridge, MA: MIT Press1986): 318–362.
56.
ChangW. S.NaS. J., Prediction of laser-spot-weld-shape by numerical analysis and neural networkMetall. Mater. Trans. B32B (2001): 723–731.
57.
ChangW. S.HapraS. C. C.CanaleR. P., Numerical methods for engineers (New York: McGrawHill Book Co.1988).
58.
KimD. W.KimB.ShimI. J.ParkG. T., Plasma etch prediction using genetic algorithm based polynomial neural networkSurf. Eng.20 (2004): 31–36.
59.
RumelhartD. E.HintonG. E.WilliamsR. J., Learning internal representations by back propagation errorsNature323 (1986): 533–535.
60.
LightfootM. P.BruceG. J.McPhersonN. A.WoodsK., The application of artificial networks to weld-induced deformation in ship plateWeld. J.84 (2005): 23s–26s.
61.
PalaniP. K.MuruganN., Prediction of delta ferrite content and effect of welding process parameters in claddings by FCAWInt. J. Mater. Manuf. Process.21 (2006): 431–438.
