Analysis of shrinkage on thick plate part using genetic algorithm
Abstract
Injection moulding process has been widely used to produce plastic products with
various shapes for high productivity and high volume products with low cost. However
there are several defect which can influence the quality of the moulded part. The most
common defect is shrinkage. Shrinkage causing contractions on moulded part. It is very
difficult to eliminate the shrinkage problem perfectly but it can be reduced. To overcome
this problem, Genetic Algorithm (GA) is used. Most of researchers use GA method to find
the shrinkage especially for thin part but hardly to found researches that use GA method for
optimising a thick plate part. A thick plate part different with the thin plate part not only the
thickness but also the weight of the part. This study aims to determine the appropriate
injection moulding parameters by simulation of Autodesk Moldflow Insight 2012 for
experimental works, determine the significant parameters that affected the shrinkage of the
thick plate part in the injection moulding process and determining an optimum shrinkage on
the thick plate parts by using GA method. This study is involving simulation, optimisation
and experimental work. Using GA, the shrinkage value of thick plate part in parallel
direction is improved by 39.1% and the shrinkage in normal direction is improved by
17.2% through simulation. Meanwhile, the shrinkage value in both parallel and normal
direction is improved by 7.1% and 13.1%, respectively through experimental. Validation
conducted gives the shrinkage in both parallel and normal direction improved by 8.4% and
18.5% through simulation and the shrinkage in both parallel and normal direction improved
by 14.5% and 7% through experimental.