Adaptive neuro fuzzy inference system modelling of multi-objective optimisation of electrical discharge machining process using single-wall carbon nanotubes

Abstract

Electrical discharge machining (EDM) has new avenues for finishing of hard and brittle materials with nano surface finish, high tolerance and accuracy. The single-wall carbon nanotube (SWCNT) is mixed with kerosene dielectric fluid in die sinking EDM process. The analysis of surface characteristics like surface roughness and metal removal rate of AISID2 tool steel materials were carried out and an excellent machined nano finish can be obtained by setting the machining parameters at optimum level. This study deals with modelling of surface roughness with SWCNT-based EDM using adaptive neuro fuzzy inference system approach. The first-order Sugeno-type fuzzy interference modelling has been used to predict the output parameters and compared with experimental values. The R2 value of regression model for with CNT is 0.706 and for without CNT is 0.652. The high R2 indicate that better model fit the data very well using CNT-based machining. The proposed model can also be used for estimating surface roughness and metal removal rate online.