Researchers develop optimisation framework for laser Directed Energy Deposition
June 4, 2025
Researchers from the Department of Materials Science and Engineering, University of Toronto, Canada, recently published a study in Additive Manufacturing focused on the use of machine learning frameworks for optimising metal laser-based Directed Energy Deposition (DED).
TRUSTED CONTENT. TARGETED AUDIENCE
Advertise with Metal AM and access a global base of 50,000+ AM professionals
| Contact us |
The identification of optimal process parameters in Additive Manufacturing can be difficult, explain the authors. Despite advances in simulations, process optimisation for specific materials and geometries is developed through a sequential and time-consuming trial-and-error approach and can lack the versatility to address multiple optimisation objectives. Machine learning (ML) provides a powerful tool to accelerate the optimisation process, but most current studies focus on a simple single-track build, which can’t accurately translate to manufacturing the bulk 3D components for engineering applications.
In the work, ‘Accurate inverse process optimization framework in laser directed energy deposition’, the researchers developed AIDED (Accurate Inverse process optimisation framework in laser Directed Energy Deposition), based on machine learning models and a genetic algorithm.
Using this framework, the researchers were able to demonstrate:
- Accurate prediction of the area of single-track melt pool (R2 score 0.995), the tilt angle of multi-track melt pool (R2 score 0.969), and the cross-sectional geometries of multi-layer melt pool (1.75 % and 12.04 % errors in width and height, respectively) directly from process parameters
- Determination of appropriate hatch spacing and layer thickness for fabricating fully dense (density > 99.9 %) multi-track and multi-layer builds
- Inverse identification of optimal process parameters directly from customisable application objectives within 1–3 hours
The study also went on to validate the effectiveness of the AIDED experimentally by solving a multi-objective optimisation problem to identify the optimal process parameters for achieving high build speeds with small effective track widths.
The transferability of the framework from stainless steel to pure nickel using a small amount of additional data on pure nickel. With such transferability in AIDED, the researchers aimed to develop a new way to aid the process optimisation of the laser-based AM processes that can apply to a wide range of materials.
‘Accurate inverse process optimization framework in laser directed energy deposition’ is available here.
