Genetic Algorithm for Optimization using Genealogy Plots to Investigate the Influences of Heat Input Ratio on Burn – Through Prevention at a GTAW Process
Queeneth Adesuwa Omoyibo–Kingsley1, Achebo J. I.2
1Queeneth Adesuwa Omoyibo – Kingsley, Ph.D student, Department of Production Engineering, University of Benin (UNIBEN), Benin City, Edo State Nigeria.
2Achebo J.I., Department of Production Engineering, Faculty of Engineering, University of Benin (UNIBEN), Benin City, Edo State Nigeria.
Manuscript received on May 02, 2017. | Revised Manuscript received on May 21, 2017. | Manuscript published on June 20, 2017.| PP: 6-9 | Volume-4 Issue-6, June 2017 | Retrieval Number: F0754054617/2017©BEIESP
Open Access | Ethics and Policies | Cite
© The Authors. Published By: Blue Eyes Intelligence Engineering & Sciences Publication (BEIESP). This is an open access article under the CC BY-NC-ND license (http://creativecommons.org/licenses/by-nc-nd/4.0/)
Abstract: Gas Tungsten Arc Welding (GTAW) offers a variety of shielding gas for joining different steel plates with different thicknesses. It is a technique that enhances the service performance of welded joints using different gas flow rate values so as to have comparison between the joints made. To avoid weld burn-through, excessive heat input and too high welding current must be regulated at a GTAW process so as to control welding current and achieve a sufficient heat input that will prevent weld burn-through and produce quality welds free from spatter. During welding, escape of gas when prevented, will produce clean joints, slow cooling (allows gases to escape) and proper shielding (absorbing the gases C02 , N2 and H2 from the atmosphere). This paper investigates burn-through prevention in welds produced, using Genetic Algorithm (GA) for optimization using genealogy plots to study the influence of Gas Flow Rate (F) ,welding current and heat input ratio on burn-through in welds. Genetic Algorithm was used for optimization in order to reach optimal solution after satisfying constraints used. In order to alleviate the problem that burn -through defect poses on welds, a matrix design was developed to determine the experimental results obtained and the results were applied to the GA model for optimization in order to obtain optimal values of responses as well as optimal values of input process parameters. Genealogy plots were used to validate the model. The genetic algorithm optimization results showed that the gas flow rate should be set to 17.67 lit/min for optimum input process parameter value. Heat input ratio optimal values of 21.96KJ/min and welding current optimal value set at 167.3 amperes were established.
Keywords: Genetic Algorithm (GA), Genealogy plots, Heat input Ratio, welding current (I), Burn-through defect and GTAW process