b'Intelligent AdditiveCorrelating additive manufacturing process parameters to the microstructure Manufacturing formation enables intelligent manufacturing design and reduced partvariability.T his project sought to develop a better understanding of process parameters and to optimally model and control process parameters to intelligently manufacture materials and components. By creating parts with more uniform build properties, the completed research helps fill a gap to allow for additive manufacturing to become more mainstream in industries like the energy industry. TOTAL APPROVED AMOUNT:Previous additive manufacturing builds contained variability in the characteristics in $1,287,038 over 3 years printed parts due to poor process control. The completed work used a combination of PROJECT NUMBER:modeling and simulation with process monitoring and control to minimize variability 18P37-030 in printed properties and allow for more intelligent selection of final properties for the build part. The initial material studied was stainless steel (primarily 316L), which PRINCIPAL INVESTIGATOR:is a well-characterized high-temperature material used in nuclear and other energy Alexander Abboud applications. The researchers developed a laser control system based on feedback CO-INVESTIGATORS: from an optical camera monitoring the melt pool. Three models were developeda Michael McMurtrey, INL computational fluid dynamics model, a phase-field model, and a FEM. These Byron Pipes, Purdue University models feed information between them for accurate thermal history, microstructure Marco Schoen, Idaho State University development, and melt pool behavior. Significant progress was made in monitoring Min Long, Boise State University the effects of process parameters on final print properties, which will provide validation for the models and an increased understanding in the relation between process parameters and printed properties. The melt pool proportionalintegralderivative controller effectively maintained consistent build dimensions. An infrared camera was used to measure the temperature and should be useful in future efforts. Improving the melt pool controller with machine learning methods was also studied.TALENT PIPELINE:Andy Lau, student at Boise State UniversityAnthony Favaloro, postdoc at Purdue UniversityAsa Monson, student at Idaho State UniversityEduardo Barocio, student at Boise State UniversityGolam Jama, student at Idaho State UniversityKanan Chowdhury, student at Idaho State University(a) The melt pool controller adjusts laser power feedback to maintain melt pool size, (b) and (c) multi-scale modeling and simulation using MOOSE, FEMs, and computational fluid dynamics (CFD), and (d) test parts validating models and demonstrating proof-of-concept.94'