Determining process parameters for optimum weld quality in submerged arc welding process of mild steel using a hybrid Fuzzy-MABAC approach

• Published in: International journal on interactive design and manufacturing Vol. 18; no. 6; pp. 4295 - 4314
• Main Authors: Biswas, Tapas Kumar, Chaki, Sudipto, Bose, Dipankar
• Format: Journal Article
• Language: English
• Published: Paris Springer Paris 01.08.2024; Springer Nature B.V
• Subjects: Algorithms; Approximation; CAE) and Design; Complexity; Computer-Aided Engineering (CAD; Content analysis; Controllability; Design of experiments; Electric potential; Electrodes; Electronics and Microelectronics; Engineering; Engineering Design; Fuzzy control; Fuzzy sets; Genetic algorithms; Geometry; Heat affected zone; Impact strength; Industrial Design; Instrumentation; Literature reviews; Low carbon steels; Mechanical Engineering; Multiple criteria decision making; Multiple criterion; Optimization; Original Article; Orthogonal arrays; Process parameters; Resource utilization; Sensitivity analysis; Statistical methods; Steel; Submerged arc welding; Taguchi methods; Tensile strength; Voltage; Welding current; Welding fluxes; Welding parameters; Submerged arc welding; Multi-attributive border approximation area comparison (MABAC) method; Optimisation; Fuzzy method
• ISSN: 1955-2513; 1955-2505
• DOI: 10.1007/s12008-024-01997-7

The submerged arc welding (SAW) process involves a complex relationship between controllable input parameters and measurable output characteristics. Several research works based on statistical methods, nature-inspired algorithms, and Multicriteria decision-making methods have been carried out to optimise submerged arc welding process parameters for effective utilization of resources and improved weld quality. Past researches suffer from limitations like the tendency to get trapped into local minima, lack of repeatability of optimised output, subjective nature of the methods, mathematical complexity, rank reversal, or bias in determining user-dependent weights. In the present work, a hybrid Fuzzy-MABAC approach has been employed that can overcome the difficulties of the previous approaches to optimise submerged arc welding process parameters. The 27 numbers of experiments have been conducted based on 3 level 5 factor L 27 Taguchi orthogonal array where arc voltage, welding current, welding speed, electrode stick out, and reuse of flux are considered as input parameters and weld quality has been assessed through the combined effect of measured output characteristics such as, weld hardness (H), weld bead width (BW), and heat affected zone (HAZ) respectively. The triangular Fuzzy method has been used to determine the weights of output parameters. Further, Multi-Attributive Border Approximation Area Comparison (MABAC) utilized the weights to determine optimised SAW output. The highest overall score of 0.6144 indicates optimum weld quality with 84 HRB, 11.1 mm, and 11.4 mm for H, BW, and HAZ respectively. Corresponding input parameters are 175A, 26 V, 1.2 m/min, 45 mm, and 30% for welding current, arc voltage, welding speed, electrode stick-out length, and reuse of flux respectively. Optimised weld quality has been achieved by moderate current, low arc voltage, moderate welding speed, adequate electrode stick out, and percentage reuse of flux. Welding voltage is found to affect the overall weld quality most significantly. The sensitivity analysis has been carried out with adequate details. The performance of the proposed Fuzzy-MABAC method corroborates well with the popular Fuzzy-TOPSIS method. Graphical abstract