Predictive modeling of dimensional accuracies in 3D printing using artificial neural network

Subramonian, Sivarao and Kumaran, K. and Dharsyanth, R. and Md Ali, Mohd Amran and Salleh, Mohd Shukor and Pujari, Satish and Ramasamy, Devarajan and Vatesh, Umesh Kumar and Mahdi Al-Obaidi, Abdulkareem Sh and Ramesh, S. and Lee, Kit Yee Sara (2023) Predictive modeling of dimensional accuracies in 3D printing using artificial neural network. Journal Of Engineering Science And Technology, 18 (4). pp. 2148-2160. ISSN 1823-4690

Text 0061718122023.PDF Download (683kB)

Abstract

Additive manufacturing, particularly Fused Deposition Modeling (FDM) using three-dimensional (3D) printing, has revolutionized the manufacturing industry by offering design flexibility, customization options, affordability, and high printing speed. However, improper selection of process parameters in FDM can lead to suboptimal surface efficiency, defective mechanical properties, increased waste, and higher production costs. In this research, an Artificial Neural Network (ANN) model was developed to optimize dimensional properties in FDM by considering control factors such as layer thickness, orientation, raster angle, raster width, and air gap. Experimental data consisting of 27 sets of control parameters and corresponding dimensional outputs were used to train and validate the ANN model. The ANN model was developed using MATLAB software, employing training functions and learning algorithms to optimize the neural network architecture. The optimized ANN structure comprised 15 neurons and 2 layers, and it demonstrated accurate prediction of dimensional properties with percentage errors ranging from 0.01% to 25.49% for length, less than 10% for weight, and less than 4% for thickness. The mean absolute percentage error (MAPE) and root mean square error (RMSE) were used to quantify the errors, indicating the effectiveness of the ANN model in predicting dimensional properties. The results highlight the potential of ANN in optimizing FDM process parameters for improved dimensional accuracy. The ANN model provides a reliable tool for manufacturers to predict and optimize the length, weight, and thickness of 3D-printed components, leading to enhanced product quality and reduced production costs. The developed ANN model can be further extended to consider other parameters and optimize various aspects of the additive manufacturing process.

Item Type:	Article
Uncontrolled Keywords:	Additive manufacturing, Artificial neutral network, Dimensional accuracy, Fused deposition modelling, Predictive modelling
Divisions:	Faculty of Manufacturing Engineering
Depositing User:	Norfaradilla Idayu Ab. Ghafar
Date Deposited:	23 Jul 2024 09:11
Last Modified:	23 Jul 2024 09:11
URI:	http://eprints.utem.edu.my/id/eprint/27419
Statistic Details:	View Download Statistic

Actions (login required)

View Item