Description
Directed Energy Deposition is a metal additive manufacturing process with applications in demanding industries with stringent requirements. Quality assurance is critical to make Directed Energy Deposition reliable and repeatable. In this work, a set-up to collect real-time coaxial NIR infrared images of the melt pool is presented and an experimental plan is conducted. The collected NIR data is used in conjunction with expert knowledge of the process to build a data-driven model that flags irregular melt pools that could lead to defects.The MiCLAD machine, designed at VUB (Belgium), is equipped with an in-situ hyperspectral NIR camera that is monitoring the melt pool during the building process. This camera captures the emitted light from the hot melt pool at 25 different wavelengths.
This high-dimensional data is used as training data for a Convolutional Neural Network.
A study is conducted to compare the efficiency of the model depending on the wavelength(s) that are considered in the training dataset. The correlation performance is be assessed by applying the model on a test set partitioned from the original dataset. This virtual sensing approach is a stepping stone for future condition monitoring of the Directed Energy Deposition process.
| Periode | 29 mrt. 2023 |
|---|---|
| Evenementstitel | Measuring by Light |
| Evenementstype | Conference |
| Locatie | Delft, NetherlandsToon op kaart |