Page 37 - IJAMD-2-3

International Journal of AI for
Materials and Design

ORIGINAL RESEARCH ARTICLE
A biomimetic machine learning approach for

predicting the mechanical properties of additive
friction stir deposited aluminum alloy-based

walled structures

1,2
Akshansh Mishra *
1 Department of Chemistry, Material and Chemical Engineering, School of Industrial and
Information Engineering, Politecnico Di Milano, Milan, Italy
2 Computational Materials Research Group, AI Fab Lab, Maharajganj, Uttar Pradesh, India

Abstract

Additive friction stir deposition (AFSD) is a solid-state manufacturing technique
capable of producing high-strength, defect-free metal components. The complexity
of its process parameters has driven growing interest in machine learning (ML)
for improved predictive accuracy and process control. This study presents a novel
biomimetic ML approach to predict the mechanical properties of AFSD-fabricated
aluminum alloy-walled structures. The methodology integrates numerical modeling
of the AFSD process with genetic algorithm (GA)-optimized ML models to predict von
Mises stress and logarithmic strain. Finite element analysis was employed to simulate
*Corresponding author: the AFSD process for five aluminum alloys: AA2024, AA5083, AA5086, AA7075, and
Akshansh Mishra AA6061, capturing the complex thermal and mechanical interactions involved.
(akshansh.mishra@mail.polimi.it)
A dataset of 200 samples was generated from these simulations. Decision tree and
Citation: Mishra A. A biomimetic random forest (RF) regression models, optimized using GAs, were developed to predict
machine learning approach for
predicting the mechanical properties key mechanical properties. The RF model demonstrated superior performance,
of additive friction stir deposited achieving R² values of 0.9676 for von Mises stress and 0.7201 for logarithmic strain.
aluminum alloy-based walled This innovative approach provides a robust tool for understanding and optimizing
structures. Int J AI Mater Design.
2025;2(3):31-44. the AFSD process across a range of aluminum alloys, offering valuable insights into
doi: 10.36922/ijamd.5014 material behavior under various process parameters.
Received: September 30, 2024
Revised: March 26, 2025 Keywords: Additive friction stir deposition; Additive manufacturing; Machine learning;
Hybrid algorithms
Accepted: March 28, 2025
Published online: July 9, 2025
Copyright: © 2025 Author(s).
This is an Open-Access article 1. Introduction
distributed under the terms of the
Creative Commons Attribution Additive friction stir deposition (AFSD) is a friction stir-based additive manufacturing
License, permitting distribution, process that enables the layer-by-layer deposition of materials using a combination of
and reproduction in any medium, feedstock, substrate, and specialized tooling. Derived from the principles of friction
1-5
provided the original work is
properly cited. stir welding, AFSD relies on intense thermoplastic deformation rather than melting,
resulting in a refined, equiaxed microstructure in the final product.
Publisher’s Note: AccScience
Publishing remains neutral with Friction stir-based additive manufacturing technologies include three primary
regard to jurisdictional claims in
published maps and institutional variants: friction surface deposition additive manufacturing, friction extrusion additive
affiliations. manufacturing, and AFSD. In friction surface deposition additive manufacturing,

Volume 2 Issue 3 (2025) 31 doi: 10.36922/ijamd.5014