Metal Additive Manufacturing Process-Structure-Property Relational Linkages Using Gaussian Process Surrogates

Process-structure-property (PSP) relational linkages are necessary for designing, developing, and tailoring a material to exhibit desired properties, and ultimately, performance for a targeted application. Establishing PSP linkages typically involves building and testing materials from a given process until the desired properties are achieved. However, the process of generating and representing the data needed to establish these PSP linkages is often time intensive as it requires extensive experimentation and/or complex, multi-scale simulations. These PSP linkages are of particular interest for additive manufacturing (AM) processes due to the plethora of process parameters involved that are significantly more influential in the localization of microstructural morphologies, and the associated material properties, compared to those of conventional manufacturing processes. With sufficient understanding of PSP linkages, their control within AM can produce parts with previously unattainable properties, including spatial distributions of those properties at will. The present works establishes and demonstrates a Gaussian process (GP) based framework to emulate experiments and simulations of AM PSP linkage components. The goal of this framework is to construct predictive surrogate models that significantly reduce the time to sample PSP link-ages while maintaining a high predictive accuracy. The framework will consist of three primary components: 1) a multi-fidelity GP linking process parameters to printability and melt pool characteristics, 2) a multi-output GP linking the melt pool to microstructure statistics, and 3) a functional GP relating microstructure statistics to the mechanical properties. The application of the described framework will demonstrate a novel linkage directly from process parameters to microstructure properties. The results demonstrate approximately 95% accuracy in predicting mechanical proper-ties for previously unseen process parameters that have been propagated through the PSP linkage. The use of GPs in the workflow limits the number of experiments/simulations needed, yields a reduced cost for acquisition of new predictions, and allows for a Bayesian treatment of the PSP linkages.

Name thesis:degree_name

Doctor of Philosophy

Level thesis:degree_level

Doctoral

Discipline thesis:degree_discipline

Industrial Engineering

Grantor

Texas A&M University

Year dc:date.issued

2022

Author dc:creator

• Saunders, Robert N

Advisors dc:contributor.advisor

• Elwany, Alaa
• Lagoudas, Dimitris

Committee members dc:contributor.committeemember

• Arroyave, Raymundo
• Michopoulos, John

• Process Modeling
• Microstructure
• Mechanical Properties
• Stainless Steel
• Gaussian Process

Language dc:language.iso

en

Handle dc:identifier.uri

https://hdl.handle.net/1969.1/197958

OAI record oai:identifier

https://oaktrust.library.tamu.edu/server/oai/request?verb=GetRecord&metadataPrefix=dim&identifier=oai%3Aoaktrust.library.tamu.edu%3A1969.1%2F197958