Design and optimization of machinability of ZnO embedded-glass fiber reinforced polymer composites with a modified white shark optimizer

This study investigates the role of a newly developed metaheuristic algorithm on machinability (cutting force) and surface roughness of nano zinc oxide embedded glass fiber reinforced polymer composites (nZnO-GFRPC). A hybrid grey theory-white shark optimizer (Grey-WSO) algorithm is developed where grey theory combines output responses (surface roughness and cutting force) into a single objective function, and white shark is used to find the optimal responses. The novelty of the developed method is the compatibility of two different varieties of machine learning algorithms into one and the combination of two different responses, i.e., cutting force and surface roughness, into a single objective function. The influence of parameters, i.e., nanoparticles amount, fiber volume fraction and feed rate, is designed by Taguchi orthogonal array and their optimization is performed by Grey-WSO. The optimal results are achieved with 1 % ZnO (Weight %), 75 mm/min feed rate and 6.031 % fiber volume fraction, respectively. The optimum cutting force and surface roughness results were 197.64 N and 1.6765 mu m, respectively. The validation of results shows that the output performance improved from 0.9414 to 0.9514, indicating the performance of the developed Grey-WSO with a 1.06% error. The developed algorithm was compared with other metaheuristics algorithms to demonstrate its potential to adopt in cutting, milling, shaping and other machining characteristics of composite materials. The results also confirm that nanoparticles amount is a highly influencing factor for surface roughness calculations.