Automated welding becomes an integral part of the modern fabrication industry because of the gains in productivity, profitability, and high safety measures. In the broad scenario of automated welding, submerged arc welding (SAW) has been expansively used because of its unique advantages of various control parameters. SAW process is extensively used in the fabrication of pipelines, off-shore structures, pressure vessels, etc. due to its inherent benefits such as excellent strength of the joint, higher metal deposition rate, high rate of production, excellent shielding effect, high melting efficiency, and automation simplicity. The mechanical strength of a weld is impacted both by the metal structure and by the reliability and form of welded joint, so an accurate method of choosing process variables and attempting to control welding parameters has become crucial in such automated applications. To accomplish this, precise relationships between process parameters and weld bead controlling parameters must be established through appropriate investigation and optimization methods. In the present work, submerged arc welding operation has been performed following central composite design (CCD) matrix. The effect of three important control parameters, viz. welding current, arc voltage, and travel speed, on ultimate tensile strength (UTS) and hardness has been analyzed using ANOVA, and the mathematical models of the same are developed. Welding current, the interaction of welding current (WC) and arc voltage, and interaction of WC and travel speed are most significant parameters affecting the UTS whereas the hardness of the welded joint is greatly influenced by interaction among welding current and travel speed. Voltage has no appreciable effect on any of the responses. To validate the predicted optimal values, confirmatory experiments have been conducted and the recorded responses are found to exist within confidence interval of 95% as per multi-response optimization.

错误:搜索内容不能为空,请输入英文关键词
错误:关键词超出字数限制,请精简
高级检索

Application of Response Surface Methodology for Optimization of Submerged Arc Welding Process Parameters

  • Rimeshikha Saikia,
  • Plabon Kakoti

摘要

Automated welding becomes an integral part of the modern fabrication industry because of the gains in productivity, profitability, and high safety measures. In the broad scenario of automated welding, submerged arc welding (SAW) has been expansively used because of its unique advantages of various control parameters. SAW process is extensively used in the fabrication of pipelines, off-shore structures, pressure vessels, etc. due to its inherent benefits such as excellent strength of the joint, higher metal deposition rate, high rate of production, excellent shielding effect, high melting efficiency, and automation simplicity. The mechanical strength of a weld is impacted both by the metal structure and by the reliability and form of welded joint, so an accurate method of choosing process variables and attempting to control welding parameters has become crucial in such automated applications. To accomplish this, precise relationships between process parameters and weld bead controlling parameters must be established through appropriate investigation and optimization methods. In the present work, submerged arc welding operation has been performed following central composite design (CCD) matrix. The effect of three important control parameters, viz. welding current, arc voltage, and travel speed, on ultimate tensile strength (UTS) and hardness has been analyzed using ANOVA, and the mathematical models of the same are developed. Welding current, the interaction of welding current (WC) and arc voltage, and interaction of WC and travel speed are most significant parameters affecting the UTS whereas the hardness of the welded joint is greatly influenced by interaction among welding current and travel speed. Voltage has no appreciable effect on any of the responses. To validate the predicted optimal values, confirmatory experiments have been conducted and the recorded responses are found to exist within confidence interval of 95% as per multi-response optimization.