Integration of Artificial Neural Network and Taguchi Method for Prediction and Minimisation of Thick-Walled Polypropylene Gear Shrinkage

Articles in Press

Document Type : Special Issue: Mechanics of Advanced Fiber Reinforced Composite Structures

Authors

1 Department of Mechanical Engineering, PDPM Indian Institute of Information Technology Design & manufacturing Jabalpur Dumna Airport Road, Dumna – 482005, India

2 Department of Mechanical Engineering, Dr B R Ambedkar National Institute of Technology Jalandhar

Abstract

The main aim of this research intended to optimise the injection moulding process parameters in order to mitigate the shrinkage of polypropylene (PP) spur gears. The methodology was used that integrated experimental approaches with artificial neural networks (ANN) and Taguchi methods to determine the most optimal combination of injection moulding parameters. The experimental data was used to create an ANN model using Matlab software that accurately predicts unseen data with a variation of less than 5%. The ANN model was then used to predict shrinkage in the context of Taguchi-based design of experiments. The investigation involved the use of Taguchi and analysis of variance techniques, determining that cooling time is the most important and relevant parameter. This is followed by packing time and melt temperature. The analysis revealed that the gears saw the least amount of shrinkage when the moulding was carried out using the optimal combination of injection moulding parameters.

Keywords

Main Subjects

Mechanics of Advanced Composite Structures

Articles in Press, Accepted Manuscript
Available Online from 15 August 2024

Files

Share

How to cite

Statistics