SBIR/STTR Award attributes
The uncertainty in predicting surface roughness and subsurface defects in as-built additively manufactured (AM) parts is hindering the wide utilization of AM for mission critical components that exhibit fatigue loading conditions due to the propagated uncertainly in crack initiation and propagation from surface defects. Thus, this work is focused on developing a new multiphysics toolset for predicting surface roughness using an ICME-based integrated computational adaptive additive manufacturing (iCAAM) suite of software with its integration of finite element simulation of AM process, machine learning tools for pattern recognition and surrogate modeling.The new developed software will be an additional module for iCAAM which will allow the integration with current development of fatigue life prediction based on damage tolerance analysis and hence enabling fast transition to DoN for flight critical applications. Throughout the project, Materials Resources LLC (MRL) in collaboration with University or Memphis (UofM) will orchestrate an experimental and modeling hybrid approach for LPBF of Ti6Al4V. The experimental work will be complimented with numerical simulations with predictions of thermal profiles. For fast predictions on a desktop computer, MRL will utilize its machine learning tools to build response surfaces that results in very fast predictions with linkages to fatigue life predictions.
