Grey Box Modelling to Predict Tensile Properties in PBF-LB

Additive Manufacturing is an upcoming technology to produce metal structures in industry as complex near net-shaped structures can be built. One commonly used method is Laser Powder Bed Fusion (PBF-LB/M), that uses lasers to melt metal powder layer by layer. Alongside advantages that come with this technology, the variety of adjustable process parameters is challenging for optimizing the process due to complex correlations of those. Additionally, the resulting mechanical properties can be challenging to optimize empirically, based on the vast multi-dimensional parameter space. To solve this problem, several Machine Learning approaches have been used in the scope of different applications. With the aim of predicting tensile properties of PBF-LB parts, three Machine Learning models (Linear Regression, XGBoost, and a Dense Neural Network) were trained on a newly created dataset. To include as much information as possible, a dimensionless metrics model was used. The models gained insights into the thermodynamic conditions of certain parameter sets. This enabled a more even distribution of different processing conditions in the dataset. Due to the small size of the dataset of only 148 data vectors, strong overfitting was observed. This could be minimized for the case of the Dense Neural Network by means of a customized architecture, strong regularization and the selection of the best model state. It was shown, that the dataset enabled sufficient predictions of yield and tensile strength with R ² values of up to 88.3 % and 86.2 %, respectively. Elongation at break turned out to be more challenging to predict with an R ² of up to 75.6 %. The results were achieved by the Dense Neural Network