Related Experiment Video
Updated: May 22, 2025
09:17
Surrogate Model Development for Digital Experiments in Welding
Published on: March 28, 2025
634
On the Prediction and Optimisation of Processing Parameters in Directed Energy Deposition of SS316L via Finite
Mehran Ghasempour-Mouziraji1,2, Daniel Afonso1,2, Ricardo Alves de Sousa1,2
1TEMA-Centre for Mechanical Technology and Automation, Department of Mechanical Engineering, University of Aveiro, 3810-193 Aveiro, Portugal.
Materials (Basel, Switzerland)
|March 13, 2025
Summary
This study optimizes Directed Energy Deposition (DED) parameters for SS316L using finite element simulation and machine learning. The integrated approach precisely controls residual stress and geometrical deviation in additive manufacturing.
Area of Science:
- Materials Science and Engineering
- Computational Materials Science
- Additive Manufacturing
Background:
- Directed Energy Deposition (DED) is a key additive manufacturing process for metals like SS316L.
- Controlling residual stress and geometrical deviation is critical for DED part quality.
- Optimization of DED processing parameters is complex and often relies on empirical methods.
Purpose of the Study:
- To develop an integrated framework combining finite element (FE) simulation and machine learning (ML) for DED process optimization.
- To identify the optimal combination of processing parameters (laser power, scanning speed, laser beam radius) for SS316L.
- To enhance precision and efficiency in controlling residual stress and displacement in DED.
Main Methods:
- Finite element (FE) simulation was employed and validated against prior research.
- A series of FE simulations were performed considering laser power, scanning speed, and laser beam radius as inputs.
- Artificial neural networks (ANNs) and a non-dominated sorting genetic algorithm (NSGA-II) were utilized for ML-based optimization.
- Confirmatory tests were conducted to validate the ML-derived results.
Main Results:
- The study successfully integrated FE simulation and ML to optimize DED processing parameters for SS316L.
- The developed methodology demonstrated enhanced precision and efficiency in managing residual stress and displacement.
- ML-predicted optimal parameters were validated through experimental testing, confirming their efficacy.
Conclusions:
- The combination of FE simulation and ML offers a powerful and novel approach for DED process optimization.
- This integrated methodology provides a practical solution for minimizing residual stress and geometrical deviations.
- The findings contribute to advancing additive manufacturing technologies through improved process control and material performance.
