A study on machinability evaluation of Al-Gr-B4C MMC using response surface methodology-based desirability analysis and artificial neural network technique. (2019)

• Record Type: Journal Article
• Authors: Ponnuvel, S.; Senthilkumar, N.
• Is Part Of: International journal of rapid manufacturing. Volume 8:Number 1/2(2019)
• Page Start: 95
• Page End: 122
• Publication Date: 2019
• Subjects: metal matrix composite -- wire-cut electrical discharge machining -- WEDM -- response surface methodology -- RSM -- desirability analysis -- artificial neural network -- ANN; Manufacturing processes -- Periodicals; 670.4205
• Languages: English
• ISSNs: 1757-8817

Abstract

In this work, machinability behaviour of aluminium-graphite-boron carbide metal matrix composite is performed during wire-cut electrical discharge machining (WEDM) process. Experiments were designed using central composite-face centred design of response surface methodology (RSM) and with the application of desirability function multiple quality characteristics viz., kerf width, surface roughness and material removal rate (MRR) were optimised simultaneously. Input parameters gap voltage, pulse ON-time, pulse OFF-time and % reinforcement of boron carbide particles in the aluminium matrix are considered. The optimised machining condition obtained is a gap voltage of 150 V, pulse ON-time of 124.56 ms, pulse OFF-time of 48.03 ms and 2.5% reinforcement of boron carbide. From the experimental values, it is observed that better output responses are achieved with lower reinforcement of boron carbide. Second order regression models are developed individually for the output responses. An artificial neural network model is developed to predict the output responses, results obtained show that a better prediction can be achieved through artificial intelligent technique.