Ingenieurwissenschaften und zugeordnete Tätigkeiten
Filtern
Erscheinungsjahr
- 2023 (2)
Dokumenttyp
- Beitrag zu einem Tagungsband (1)
- Vortrag (1)
Sprache
- Englisch (2)
Referierte Publikation
- nein (2)
Schlagworte
- Digital Twin (2) (entfernen)
Organisationseinheit der BAM
Eingeladener Vortrag
- nein (1)
The industrialization of AM is only possible by creating synergy with the tools of Industry 4.0. The system technology of Powder Bed Fusion with Laser beam of Metals (PBF-LB/M) reached a level of high performance in terms of process stability and material spectrum in the past years. However, the digital process chain, starting from CAD via CAM and plant-specific compila-tion of the manufacturing file exhibits media disruptions. The consequence is a loss of metadata. A uniform data scheme of simulation for Design for Additive Manufacturing (DfAM), the PBF-LB/M process itself and quality assurance is currently not realized within industry. There is no entity in the common data flows of the process chains, that enables the integration of these functionalities. As part of the creation of a digital quality infrastructure in the QI-Digital pro-ject, an integration of the CAD/CAM chain is being established. The outcome is a file in an advanced commercially available format which includes all simula-tions and manufacturing instructions. The information depth of this file extends to the level of the scan vectors and allows the automatic optimization and holis-tic documentation. In addition, the KPI for the economic analysis are generated by compressing information into a unique file combined with the application of a digital twin. The implementation and advantages of this solution are demon-strated in a case study on a multi-laser PBF-LB/M system. A build job contain-ing a challenging geometry is thermally simulated, optimized, and manufac-tured. To verify its suitability for an Additive Manufacturing Service Platform (AMSP), the identical production file is transferred to a PBF-LB/M system of another manufacturer. Finally, the achieved quality level of the build job is evaluated via 3D scanning. This evaluation is carried out in the identical entity of the production file to highlight the versatility of this format and to integrate quality assurance data.
In recent years, the use of simulation-based digital twins for monitoring and assessment of complex mechanical systems has greatly expanded. Their potential to increase the information obtained from limited data makes them an invaluable tool for a broad range of real-world applications. Nonetheless, there usually exists a discrepancy between the predicted response and the measurements of the system once built. One of the main contributors to this difference in addition to miscalibrated model parameters is the model error. Quantifying this socalled model bias (as well as proper values for the model parameters) is critical for the reliable performance of digital twins. Model bias identification is ultimately an inverse problem where information from measurements is used to update the original model. Bayesian formulations can tackle this task. Including the model bias as a parameter to be inferred enables the use of a Bayesian framework to obtain a probability distribution that represents the uncertainty between the measurements and the model. Simultaneously, this procedure can be combined with a classic parameter updating scheme to account for the trainable parameters in the original model. This study evaluates the effectiveness of different model bias identification approaches based on Bayesian inference methods. This includes more classical approaches such as direct parameter estimation using MCMC in a Bayesian setup, as well as more recent proposals such as stat- FEM or orthogonal Gaussian Processes. Their potential use in digital twins, generalization capabilities, and computational cost is extensively analyzed.