Experimental investigation and ANFIS-based modelling of effect of process parameters on friction stir spot welding of Al 6061-T6

ABSTRACT

This research presents the behaviour of different friction stir spot welding (FSSW) process parameters on tensile-shear load (TSL) of weld joints. There is a need to optimise the parameters and develop a prediction model in order to ease and smoothen the work flow. Three parameters with three levels considered in this research were tool rotation speed (rpm), dwell time (s) and plunge rate (mm/min). The influence of these factors on tensile shear strength was of definite interest in this research. A statistical practice known as Taguchi was performed to analyse and determine the optimum combination of process parameters as 1500 rpm, 18 s and 20 mm/min, respectively. Additionally, the paper presents an adaptive neuro fuzzy inference system (ANFIS) model to estimate the output at various arrangements of input parameters. The outcomes showed that the ANFIS model accurately predicted the output and there was a minute difference between experimentally obtained values and predicted values of the response.

KEYWORDS:

• Friction stir spot welding (FSSW)
• design of experiment (DOE)
• Taguchi
• adaptive neuro fuzzy inference system (ANFIS)

Disclosure statement

Both authors have revised and ratified the final paper, and no funds or grants were obtained during this research investigation.