In response to NASA’s topic T12.02 of “Extensible Modeling of Metallurgical Additive Manufacturing Processes”, Sentient proposes to incorporate its DigitalClone technique to develop a multiscale and multiphysics computational modeling suite to predict comprehensive outcomes from AM building processes, including geometrical accuracy, and resulting microstructure and defects. Figure 1 shows the proposed framework for the multiscale modeling suite. The process model will first predict the microscale thermal evolution in respect of various parameters. The temperature results will feed a subsequent macroscale model for prediction of stress and distortion at part scale. Moreover, the predicted thermal history and distribution will feed subsequent microstructure model to further predict the micro-scale features including grain morphology and porosity. The proposed computational modeling framework allows a comprehensive prediction and understanding of the metal AM process at multiple levels.
In Phase I, Sentient will upgrade and demonstrate DigitalClone’s capability to integrate process-microstructure simulation for metal AM process. Specifically, selective laser melting of IN 718 alloy will be used for development and demonstration purposes in Phase I. AM coupons with different geometries will be fabricated by Selective Laser Melting (SLM) at different parameters. DigitalClone will be used to simulate all different scenarios of coupons made from IN718 alloys, and predict temperature, stress, part distortion, and grain structure. Materials characterization will be performed on the coupons to examine geometrical accuracy, microstructure, residual stress, all of which will be used to validate the DigitalClone model. In Phase II, different materials and AM platforms and more complex geometrical components will be tested for model validation. Additionally, close-loop optimization framework will be explored for improving geometrical design and microstructure features.
A successful completion of this project will lead to a robust AM modeling suite that provides accurate prediction of dimensional accuracy, microstructure, and defects in AM process. The proposed modeling suite will significantly reduce the uncertainty and conservatism in design of new AM components and processes. NASA would directly benefit from this software via virtually pre-testing the new AM component design, process effects and part quality.
The proposed modeling software will benefit several other industries incorporating AM technique, including aerospace, medical device, automotive industries. This will not only allow customers virtually evaluating the AM part qualities, more importantly, it will provide the “best solution” for customer in respect of optimizing AM design, selecting process and materials, increasing performance, reliability and durability, and reducing cost of operation the process.