Modeling and parametric optimization of laser powder bed fusion 3D printing technique using artificial neural network for enhancing dimensional accuracy. (2022)

• Record Type: Journal Article
• Authors: Phadke, Nikhil; Raj, Ratnesh; Kumar Srivastava, Ashish; Dwivedi, Suryank; Rai Dixit, Amit
• Is Part Of: Materials today. Volume 56:Part 2(2022)
• Page Start: 873
• Page End: 878
• Publication Date: 2022
• Subjects: Laser powder bed fusion (LPBF) -- 3D printing -- Artificial Neural Networks -- Mean Square Error -- Coefficient of correlation (R) -- Neurons; Materials science -- Congresses -- Periodicals; 620.1
• DOI: 10.1016/j.matpr.2022.02.523
• Languages: English
• ISSNs: 2214-7853

Abstract

Abstract: In this study, Artificial Neural Networks (ANN) is used to establish a relationship between the process and output parameters of the Laser Powder-based fusion process for AlSi10Mg alloy. Initially, data is collected through multiple experiments designed under response surface methodology, and then this data is used to Train, Test, and Validate a neural network that creates a simulation of the Laser Powder-based fusion process. As the process is expensive, printing samples for the sake of testing is not affordable. This is where this study comes in as the user will get the output values for the required input parameters with high accuracy, which enables the user to test and optimize the process without actually printing.