|
CSME 2021/04
Volume 42 No.2
:
227-234
|
|
Application of Regression Models in Multi-Objective Optimization of FCAW Process Variables on Volume of Austenitic Stainless-Steel Clad Layers
M.Sowrirajan a, S.Vijayan b and M.Arulraj a
aDepartment of Mechanical Engineering, Coimbatore Institute of Engineering and Technology, Coimbatore -641109, Tamilnadu, INDIA.
bDepartment of Mechanical Engineering, National Institute of Technology, Tiruchirappalli-620015. Tamilnadu, INDIA
|
Abstract:
Metal cladding is a process of depositing a thick layer of material over another material using a suitable welding process to preserve the material from corro-sion problems. Cost estimation for producing the clad-ding with desired quality is essential in fabrication industries, which includes the cost of consumable filler material. Analysis on volume of metal deposited during cladding process could provide necessary knowledge about consumption of filler wire and thereby the cost of consumables. In this work, an attempt was made to perform multi-criteria optimiza-tion for depositing a heat resistant layer over a material used in boiler construction. Therefore, low thermal conductivity 316L grade of austenitic stainless steel was surfaced over IS:2062 structural steel plates using FCAW process. Rotatable central composite design for five factors and five levels was used to perform the experiments. Mathematical models were developed for the prediction of volume of reinforcement and volume of penetration and tested for adequacy with the help of ANOVA technique. Multi-objective constrained optimization was carried out using RSM and genetic algorithm tool to yield best optimum set of process variables for the responses of interest. Optimum settings and developed models were validated by good agreement shown during conformity test experiments. The findings have wide industrial applications in the field of surfacing.
|
|
Keywords: Cladding, Stainless steel. Volume of reinforcement. Volume of penetration, Optimization, Genetic algorithm, Response surface methodology
|
Download PDF
|
|
*Corresponding author; e-mail:
|
©
2021
CSME , ISSN 0257-9731
|