Multi-objective Optimization of Parameters in Abrasive Water Jet Machining of Carbon-Glass Fibre-Reinforced Hybrid Composites

• Published in: Journal of the Institution of Engineers (India): Series D Vol. 100; no. 1; pp. 55 - 66
• Main Authors: Rao, V. Durga Prasada, Mrudula, M., Geethika, V. Navya
• Format: Journal Article
• Language: English
• Published: New Delhi Springer India 05-04-2019; Springer Nature B.V
• Subjects: Abrasive machining; Carbon; Carbon fiber reinforced plastics; Classification; Cutting parameters; Cutting speed; Design factors; Engineering; Epoxy resins; Feed rate; Fiber composites; Fiber reinforced polymers; Genetic algorithms; Glass fiber reinforced plastics; Hybrid composites; Hydraulic jets; Kerf; Material removal rate (machining); Mathematical analysis; Multiple objective analysis; Optimization; Original Contribution; Polymer matrix composites; Polymers; Production methods; Regression analysis; Sorting algorithms; Surface roughness; CGFRP; Surface roughness; Kerf width; NSGA; MRR; AWJM
• ISSN: 2250-2122; 2250-2130
• DOI: 10.1007/s40033-019-00181-6

In the present work, first an attempt has been made to prepare carbon, glass and carbon-glass fibre-reinforced polymer composites, and later, machining of these composites is done on abrasive water jet machine (AWJM) to compare and optimize the machining parameters. Actually, the carbon fibre-reinforced polymer (CFRP) composites, glass fibre-reinforced polymer (GFRP) composites and carbon-glass fibre-reinforced polymer (CGFRP) composites are prepared through vacuum bagging process by using epoxy resin as the polymer matrix. The machining experiments are conducted to analyse the effects of the predominant machining parameters, i.e. cutting speed rate, feed rate and stand-off distance on the required machining characteristics, i.e. surface roughness (Ra), kerf top width (kw) and material removal rate (MRR). The range of values of each parameter is set at three different levels, and Taguchi’s L
9
orthogonal array is used to design factors so that all the interactions between the response variables and machining variables can be investigated. Based on the experimental values, second-order regression equations are fitted between each of the response parameters and the machining parameters using Minitab 18 software. The equations are then optimized by defining the three equations of Ra, kw and MRR as the three objectives of a multi-objective optimization problem (MOOP) using a multi-objective optimization algorithm called Elitist Non-dominated Sorting Genetic Algorithm (NSGA-II). Single best compromise solutions with respect to the MOOPs of GFRP, CFRP and CGFRP composites are also determined from the Pareto optimal solutions obtained by NSGA-II. Finally, confirmation tests are conducted on specimens of GFRP, CFRP and CGFRP composites machined at their corresponding optimum parameters given by the GA. It is observed that the optimum values of Ra, kw and MRR of all the optimization problems are closer to the corresponding experimental values of confirmation tests.