An improved grey wolf optimiser sine cosine algorithm for minimisation of injection moulding shrinkage

Abstract

The injection moulding process in plastic manufacturing parts is widely used and the products can be seen anywhere as daily use items. This process includes a big scale of production. This sometimes leads to defects that affect the quality of the products. As a result, the production is inefficient, time-consuming, and costly. However, one of the solutions that have been discovered is the fact that hybridisation improves product quality, especially in minimising shrinkage defect at a thick plate part by providing the best parameter setting. For an excellent performance of the injection moulding process, it is crucial to have an optimum set of parameters and this study considered melt temperature (oC), mould temperature (oC), cooling time(s), and packing pressure (MPa) as a set of parameters. In this study, an improved hybridisation technique of Grey Wolf Optimiser Sine Cosine Algorithm (GWOSCA) was developed to estimate optimal parameter settings so that the value of shrinkage at the thick plate could be minimised. The improved GWOSCA was made to enhance the searching strategy of GWOSCA by increasing the movement of direction and speed while sharing information among the alpha, beta, and delta to find the optimum value. The simulation and improved results from GWOSCSA were compared and validated by using experimental work of percentage error, regression model, and analysis of variance (ANOVA). It showed that the improved GWOSCA could minimise the shrinkage at the thick plate by 0.48% at x-axis and 0.35% at y-axis in contrast with the simulation result, which was only 0.58% at x-axis and 0.60% at y-axis in this study. Eventually, the improved GWOSCA optimisation technique significantly showed that it could minimise the values of shrinkage in the injection moulding process for manufacturing fields.