Research has shown that fatigue failure is a major concerned for manufacturers, as it has been recorded that a good
percentage of manufactured products and structures, especially those with welded joints, tend to fail as a result of being
subjected to loads, often beyond their designed capacity. This study explores the application of optimization techniques such
as Response Surface Methodology (RSM) and Genetic Algorithm (GA) in determining the optimal Impact Strength, tensile
strength and Fatigue Life of a Gas tungsten arc welded plate with the aim of ascertaining the optimal fatigue life and strength
of the weld. With the application of both techniques, this study obtained the most adequate optimal process parameter with
the GA recording the most accurate performance. The RSM recorded optimal Impact Strength, tensile strength and Fatigue
Life values of 576.609N/mm2, 491.462N/mm2 and 288306cycles respectively, while the GA recorded optimal impact
strength, tensile strength, and fatigue life values of 587.25N/mm2, 489.81N/mm2 and 299635.0 respectively at the 119
iteration. Confirmatory test performed using the optimal values revealed that the GA technique had the most accurate
performance with a percentage error of 3% compared to the RSM results which recorded an error of 11%.
Keywords: Fatigue Life, Tensile Strength, Impact Strength, Genetic Algorithm
Key words: Fatigue Life, Tensile Strength, Impact Strength, Genetic Algorithm
|
