Abstract
The paper presents the prediction of surface quality during the electro-discharge machining (EDM) of aluminium-based alloy. The composition consists of Aluminium, copper, nickel (Al/Cu/Ni) alloy. The techniques like dimensional exponential model (DEM), response surface-based modelling (RSM), and adaptive neuro-fuzzy inference based system (ANFIS) were employed to analyse the EDM process. Response variable i.e. Surface characteristic measure in terms of roughness (Ra) is the dependent parameter. Experimental analysis was done using Taguchi’s L18 mixed OA’s i.e. orthogonal-array by considering the process parameters namely variation in the material composition (CP), pulse-on-time(TON), pulse-off- time(TOFF) and input current(IP). Analysis of variances (ANOVA) has been employed out to identify the significant process parameters and their impact over surface characteristics. Models based on the DEM and RSM were developed to analyse the surface characteristics or roughness of the Al/Cu/Ni alloy. The experimental result proves the commitment of all three approaches to represent the process effectually. The correlation coefficient (R2) between the experimental and the DEM, RSM, and ANFIS predicted roughness was 0.830626, 0.995649, and 0.999958 respectively. This value shows the dominance of the ANFIS technique over the DEM and the RSM.

This publication has 22 references indexed in Scilit: