An approach to select the optimal process parameters of laser transmission welding using firefly algorithm. (2016)

• Record Type: Journal Article
• Authors: Acherjee, Bappa; Maity, Debanjan; Kuar, Arunanshu S.
• Is Part Of: International journal of innovative computing and applications. Volume 7:Number 3(2016)
• Page Start: 163
• Page End: 178
• Publication Date: 2016
• Subjects: laser transmission welding -- LTW -- firefly algorithm -- metaheuristics -- response surface methodology -- RSM -- optimisation -- process parameters -- laser power -- welding speed -- stand-off distance -- clamp pressure -- lap shear strength -- weld seam width -- convergence rate -- particle swarm optimisation -- PSO -- differential evolution -- shuffled frog leaping; Evolutionary computation -- Periodicals; Neural networks (Computer science) -- Periodicals; Genetic programming (Computer science) -- Periodicals; Biologically-inspired computing -- Periodicals; Swarm intelligence -- Periodicals; Quantum computers -- Periodicals; 006.3
• Languages: English
• ISSNs: 1751-648X

Abstract

In the present work, laser transmission welding process is investigated and optimised using firefly algorithm, a population-based metaheuristic and bio-inspired algorithm. Laser power, welding speed, stand-off distance and clamp pressure are the laser transmission welding process parameters which control the output qualities like lap-shear strength and weld-seam width of the welded part. To select the optimal setting of process parameters, single-objective and multi-objective optimisations are performed using firefly algorithm, in which the objective functions are developed using response surface method. Results obtained by using firefly algorithm are found to be superior to those obtained using conventional optimisation technique. Firefly algorithm also found to be superior to particle swarm optimisation algorithm, differential evolutionary algorithm, and shuffled frog leaping algorithm in terms of convergence rate. It is found that firefly algorithm is an excellent optimisation tool with respect to fast convergence and locating global optima. It is also very efficient in predicting trends of multi-dependent responses.