Open AccessMulti-objective optimization and analysis for laser beam cutting of stainless steel (SS304) using hybrid statistical tools GA-RSM
Author(s) -
Amanuel Diriba Tura,
Hana Beyene Mamo,
Debela Geneti Desisa
Publication year - 2021
Publication title -
iop conference series. materials science and engineering
Language(s) - English
Resource type - Journals
eISSN - 1757-899X
pISSN - 1757-8981
DOI - 10.1088/1757-899x/1201/1/012030
Subject(s) - taguchi methods , surface roughness , genetic algorithm , response surface methodology , orthogonal array , materials science , matlab , machining , multi objective optimization , beam (structure) , laser , design of experiments , mechanical engineering , laser cutting , point (geometry) , surface finish , structural engineering , optics , computer science , composite material , mathematics , engineering , metallurgy , geometry , machine learning , statistics , physics , operating system
A laser beam machine is a non-traditional manufacturing technique that uses thermal energy to cut nearly all types of materials. The quality of laser cutting is significantly affected by process parameters. The purpose of this study is to use a genetic algorithm (GA) in conjunction with response surface approaches to improve surface roughness in laser beam cutting CO2 with a continuous wave of SS 304 stainless steel. The effects of the machining parameters, such as cutting speed, nitrogen gas pressure, and focal point location, were investigated quantitatively and optimized. The tests were carried out using the Taguchi L9 orthogonal mesh approach. Analysis of variance, main effect plots, and 3D surface plots were used to evaluate the impact of cutting settings on surface roughness. A multi-objective genetic algorithm in MATLAB was used to achieve a minimum surface roughness of 0.93746 μm, with the input parameters being 2028.712 mm/m cutting speed, 11.389 bar nitrogen pressure, and a focal point position of - 2.499 mm. The optimum results of each method were compared, as the results the response surface approach is less promising than the genetic algorithm method.