Dr. M.R. Phate
Ms. Shraddha B Toney
Vikas R. Phate
Abstract
Wire electric discharge machining i.e. WEDM is very popular non-conventional machining process. Most of the machining industries preferred WEDM as an economical and effective machining process. Along with the use of aluminium, its composite are going to be very popular and fulfill the industrial need. To maximize the use of WEDM for getting the economical and precise outcomes, analysis is very important. There are so may process parameters concern with the WEDM process, but it was really a tough job to correlate all the process parameters. Hence the parameters such as P-on time (TON), P-off time (TOFF), wire federate (WFR) and the input current (IP) has been finalized for the investigation. The new aluminium based metal matrix with graphite (5%) as a filler metal was selected for the investigation. Taguchi L27 (33 ) design of experimentation plan has been employed to carried out the experimentation. Two modelling techniques i.e. response surface method (RSM) and soft computing technique i.e. artificial neural techniques has been adopted for the analysis. From the experimental findings, it has been observed that both RSM and ANN have efficiently predicted the response with an acceptable agreement measure in terms of correlation coefficient. Feed forward back propagation neural network was employed for the analysis through ANN while the second degree response surface model was formulated for response surface method. Analysis of variance (ANOVA) was carried out to know the impact of various process parameters. But the ANN was superior to the RSM model hence recommended for the investigation of such process.
Keywords- Al/Gr/Cp5 MMC,ANN, ANOVA,RSM,WEDM.