Optimization of process parameters in plastic injection molding for minimizing the volumetric shrinkage and warpage using radial basis function (RBF) coupled with the k-fold cross validation technique

Behzad Shiroud Heidari, Amin Hedayati Moghaddam, Seyed Mohammad Davachi, Shadi Khamani, Afshar Alihosseini

Research output: Contribution to journalArticle

8 Citations (Scopus)

Abstract

In this study, a multi-objective design optimization method based on a radial basis function (RBF) model was applied to minimize the volumetric shrinkage and warpage of hip liners as an injection-molded biomedical part. The hip liners included an ultrahigh molecular weight polyethylene (UHMWPE) liner and UHMWPE reinforced with a nano-hydroxyapatite (nHA) liner. The shrinkage and warpage values of the hip liners were generated by simulation of the injection molding process using Autodesk Moldflow. The RBF model was used to build an approximate function relationship between the objectives and the process parameters. The process parameters, including mold temperature, melt temperature, injection time, packing time, packing pressure, coolant temperature, and type of liner, were surveyed to find the interaction effects of them on the shrinkage and warpage of the liners. The results indicated that the addition of nHA helps the liners to obtain more dimensional stability. The model was validated by the k-fold cross validation technique. Finally, the model revealed the optimal process conditions to achieve the minimized shrinkage and warpage simultaneously for various weights.

Original languageEnglish
Pages (from-to)481-492
Number of pages12
JournalJournal of Polymer Engineering
Volume39
Issue number5
DOIs
Publication statusPublished - 1 May 2019

Fingerprint Dive into the research topics of 'Optimization of process parameters in plastic injection molding for minimizing the volumetric shrinkage and warpage using radial basis function (RBF) coupled with the k-fold cross validation technique'. Together they form a unique fingerprint.

Cite this