Defect Classification for Additive Manufacturing with Machine Learning.

Additive manufacturing offers significant design freedom and the ability to selectively influence material properties. However, conventional processes like laser powder bed fusion for metals may result in internal defects, such as pores, which profoundly affect the mechanical characteristics of the components. The extent of this influence varies depending on the specific defect type, its size, and morphology. Furthermore, a single component may exhibit various defect types due to the manufacturing process. To investigate these occurrences with regard to other target variables, this study presents a random forest tree model capable of classifying defects in binary images derived from micrographs. Our approach demonstrates a classification accuracy of approximately 95% when distinguishing between keyhole and lack of fusion defects, as well as process pores. In contrast, unsupervised models yielded prediction accuracies below 60%. The model's accuracy in differentiating between lack of fusion and keyhole defects varies based on the manufacturing process's parameters, primarily due to the irregular shapes of keyhole defects. We provide the model alongside this paper, which can be utilized on a standard computer without the need for in situ monitoring systems during the additive manufacturing process.

Programmable Density of Laser Additive Manufactured Parts by Considering an Inverse Problem.

In this Article, the targeted adjustment of the relative density of laser additive manufactured components made of AlSi10Mg is considered. The interest in demand-oriented process parameters is steadily increasing. Thus, shorter process times and lower unit costs can be achieved with decreasing component densities. Especially when hot isostatic pressing is considered as a post-processing step. In order to be able to generate process parameters automatically, a model hypothesis is learned via artificial neural networks (ANN) for a density range from 70% to almost 100%, based on a synthetic dataset with equally distributed process parameters and a statistical test series with 256 full factorial combined instances. This allows the achievable relative density to be predicted from given process parameters. Based on the best model, a database approach and supervised training of concatenated ANNs are developed to solve the inverse parameter prediction problem for a target density. In this way, it is possible to generate a parameter prediction model for the high-dimensional result space through constraints that are shown with synthetic test data sets. The presented concatenated ANN model is able to reproduce the origin distribution. The relative density of synthetic data can be predicted with an R2-value of 0.98. The mean build rate can be increased by 12% with the formulation of a hint during the backward model training. The application of the experimental data shows increased fuzziness related to the big data gaps and a small number of instances. For practical use, this algorithm could be trained on increased data sets and can be expanded by properties such as surface quality, residual stress, or mechanical strength. With knowledge of the necessary (mechanical) properties of the components, the model can be used to generate appropriate process parameters. This way, the processing time and the amount of scrap parts can be reduced.

Adjustment of Mechanical Properties of Medium Manganese Steel Produced by Laser Powder Bed Fusion with a Subsequent Heat Treatment.

Medium manganese steels can exhibit both high strength and ductility due to transformation-induced plasticity (TRIP), caused by metastable retained austenite, which in turn can be adjusted by intercritical annealing. This study addresses the laser additive processability and mechanical properties of the third-generation advanced high strength steels (AHSS) on the basis of medium manganese steel using Laser Powder Bed Fusion (LPBF). For the investigations, an alloy with a manganese concentration of 5 wt.% was gas atomized and processed by LPBF. Intercritical annealing was subsequently performed at different temperatures (630 and 770 °C) and three annealing times (3, 10 and 60 min) to adjust the stability of the retained austenite. Higher annealing temperatures lead to lower yield strength but an increase in tensile strength due to a stronger work-hardening. The maximum elongation at fracture was approximately in the middle of the examined temperature field. The microstructure and properties of the alloy were further investigated by scanning electron microscopy (SEM), hardness measurements, X-ray diffraction (XRD), electron backscatter diffraction (EBSD) and element mapping.

Descriptors for High Throughput in Structural Materials Development.

The development of novel structural materials with increasing mechanical requirements is a very resource-intense process if conventional methods are used. While there are high-throughput methods for the development of functional materials, this is not the case for structural materials. Their mechanical properties are determined by their microstructure, so that increased sample volumes are needed. Furthermore, new short-time characterization techniques are required for individual samples which do not necessarily measure the desired material properties, but descriptors which can later be mapped on material properties. While universal micro-hardness testing is being commonly used, it is limited in its capability to measure sample volumes which contain a characteristic microstructure. We propose to use alternative and fast deformation techniques for spherical micro-samples in combination with classical characterization techniques such as XRD, DSC or micro magnetic methods, which deliver descriptors for the microstructural state.

Parameter Optimization in High-Throughput Testing for Structural Materials.

High-throughput screenings are established evaluation methods in the development of functional materials and pharmaceutical active ingredients. The transfer of this approach to the development of structural materials requires extensive adaptations. In addition to the investigation of new test procedures for the determination of material properties and the treatment of metallic materials, the design of experiments is a research focus. Based on given descriptor target values, the statistical design of experiments determines investigations and treatments for the investigation of these materials. In this context, process parameters also have to be determined, as these have a major influence on the later material properties, especially during the treatment of samples. In this article, a method is presented which determines the process parameters iteratively. The validation of the calculated process parameters takes place based on differential scanning calorimetry used as the furnace for the heat treatment of small batches and particle-oriented peening as the characterization method.