About this Abstract |
Meeting |
2025 Annual International Solid Freeform Fabrication Symposium (SFF Symp 2025)
|
Symposium
|
2025 Annual International Solid Freeform Fabrication Symposium (SFF Symp 2025)
|
Presentation Title |
Using Machine Learning-Based Process Map to Guide Evaluation of Inconel 625 Mechanical Properties in Laser Foil Printing |
Author(s) |
Yu-Hsiang Wang, Chia-Hung Hung |
On-Site Speaker (Planned) |
Yu-Hsiang Wang |
Abstract Scope |
This study presents a systematic, data-efficient methodology to guide the evaluation of mechanical properties of Inconel 625 parts fabricated using the Laser Foil Printing (LFP) process. Single-track experiments were conducted to collect melt pool depth and width data across a range of laser power and scanning speed settings. The collected data were used to train a Gradient Boosting Regression Model for constructing a predictive process map. The process map classifies melt pool modes into lacking fusion, conduction, and keyhole zones, and presents the melt pool depth-to-foil thickness (D/T) ratio using varying shades of green within the conduction region. Based on this map, multi-track multi-layer LFP experiments were conducted using various D/T ratios. The fabricated samples were analyzed for porosity, melt pool stability, and inter-layer bonding to identify the optimal LFP process parameters. |
Proceedings Inclusion? |
Planned: Post-meeting proceedings |