Development of fuzzy logic system to predict the SAW weldment shape profiles

Abstract

A fuzzy model was presented to predict the weldment shape profile of submerged arc welds (SAW) including the shape of heat affected zone (HAZ). The SAW bead-on-plates were welded by following a full factorial design matrix. The design matrix consisted of three levels of input welding process parameters. The welds were cross-sectioned and etched, and the zones were measured. A mapping technique was used to measure the various segments of the weld zones. These mapped zones were used to build a fuzzy logic model. The membership functions of the fuzzy model were chosen for the accurate prediction of the weld zone. The fuzzy model was further tested for a set of test case data. The weld zone predicted by the fuzzy logic model was compared with the experimentally obtained shape profiles and close agreement between the two was noted. The mapping technique developed for the weld zones and the fuzzy logic model can be used for on-line control of the SAW process. From the SAW fuzzy logic model an estimation of the fusion and HAZ can also be developed.

Keywords

submerged arc welding (SAW) / fuzzy-logic controller / bead height / weldment cross-sectional-area / heat affected zone (HAZ) / fuzzy model / fuzzy logic system

References

[1] Bandler W., Kohout L.J. Special properties, closures and interiors of crisp and fuzzy relations. Fuzzy Sets and Systems, 1988, 26(3): 317-332

[2] Chan B., Chandel R.S., Yang L.J., Bibly M.J. Software system for anticipating the size and shape of submerged arc welds. Journal of Material Processing Technology, 1994, 40(3/4): 249-262

[3] Chandel R.S., Bala S. Relationship between submerged arc welding parameters and weld bead size. Schweissen Schneiden, 1998, 40(2): 28-31

[4] Datta GL (2004). Metal-arc weld penetration capability-studies and analysis. Proceedings of National Seminar on Weld Integrity, Kolkata, 8–11.

[5] Dhas J.E.R., Kumanan S. ANFIS for prediction of weld bead width in a submerged arc welding process. Journal of Scientific & Industrial Research, 2007, 66(5): 335-338

[6] Kim I.S., Kwon W.H., Siores E. An investigation of a mathematical model for predicting weld bead geometry. Canadian Metallurgical Quarterly, 1996, 35(4): 385-392

[7] Kim S.I., Son J.S., Park C.W., Kim I.J., Kim H.H. An investigation into an intelligent system for predicting bead geometry in GMA welding process. Journal of Materials Processing Technology, 2002, 159: 113-118

[8] Lee C.S., Chandel R.S., Seow H.P. Effect of welding parameters on the size of heat affected zone of submerged arc welding. Material and Manufacture Process, 2010, 15(5): 649-666

[9] Lee J., Um K. A comparison in back-bead prediction of gas metal arc welding using multiple regression analysis and artificial neural networks. Optics and Laser in Engineering, 2000, 34: 149-158

[10] Mahapatra M.M., Ali M.S., Datta G.L., Pradhan B. Modelling and predicting the effects of process parameters on weldment characteristics in shielded metal arc welding. Indian Welding Journal, 2005, 38(2): 22-29

[11] Nagesh S.D., Datta G.L. Prediction of weld bead geometry and penetration in shielded metal-arc welding using artificial neural networks. Journal of Materials Processing Technology, 2002, 123: 303-312

[12] Tanaka K., Sugeno M. Stability analysis and design of fuzzy control systems. Fuzzy Sets and System, 1992, 45: 135-156

[13] Zadeh L.A. Fuzzy algorithms. Information & Control, 1968, 12: 94-102