{"id":{"repo_id":"msu","oai_identifier":"oai:d.lib.msu.edu:etd_52577"},"canonical_url":"https://search.ndltd.org/etd/msu/oai:d.lib.msu.edu:etd_52577","repository":{"repo_id":"msu","name":"Michigan State University","base_url":"https://d.lib.msu.edu/oai"},"display":{"title":"Structure Preserving, Deep Learning Methods for Fluid Moment Models","abstract":"Thesis (Ph.D.)--Michigan State University. Applied Mathematics - Doctor of Philosophy, 2025","abstract_html":"Thesis (Ph.D.)--Michigan State University. Applied Mathematics - Doctor of Philosophy, 2025","abstract_has_math":false,"creators":["Krupansky, Nicholas Anthony"],"institution":null,"degree_name":null,"degree_level":null,"degree_discipline":null,"degree_department":null,"school":null,"contributors":["Christlieb, Andrew J","O'Shea, Brian W","Qian, Jianliang","Lei, Huan"],"advisors":[],"committee_chairs":[],"committee_members":[],"year":2025,"date_issued":"2025","date_published":"2025","updated_at":"2026-11-18T18:59:09Z","subjects":["Applied mathematics","Mathematics"],"languages":["eng"],"rights":["In Copyright"],"rights_urls":[],"identifier_entries":[{"key":"dc:identifier","label":"Identifier","values":["etd:52577","local:Krupansky_grad.msu_0128D_21530"],"render_values":[{"text":"etd:52577","href":null,"code":true},{"text":"local:Krupansky_grad.msu_0128D_21530","href":null,"code":true}]}],"person_identifier_entries":[]},"links":{"outbound_url":"https://doi.org/doi:10.25335/2pa9-7w77","outbound_label":"DOI","outbound_source":"dc:identifier","record_url":"https://doi.org/doi:10.25335/2pa9-7w77","record_label":"Repository record","record_kind":"doi","record_source":"dc:identifier","download_url":null,"download_label":null,"download_source":null,"access_status":null,"embargo_until":null,"access_note":null},"source_record":{"url":"https://d.lib.msu.edu/oai?verb=GetRecord&metadataPrefix=oai_dc&identifier=oai%3Ad.lib.msu.edu%3Aetd_52577","prefix":"oai_dc"},"metadata_groups":[{"id":"people","label":"People","entries":[{"key":"dc:contributor","label":"Contributor","values":["Christlieb, Andrew J","O'Shea, Brian W","Qian, Jianliang","Lei, Huan"]},{"key":"dc:creator","label":"Author","values":["Krupansky, Nicholas Anthony"]}]},{"id":"academic_context","label":"Academic Context","entries":[{"key":"dc:date","label":"Dc Date","values":["2025"]},{"key":"dc:relation","label":"Dc Relation","values":["Electronic Theses & Dissertations"]},{"key":"dc:type","label":"Dc Type","values":["text","Electronic dissertations"]}]},{"id":"subjects_keywords","label":"Subjects and Keywords","entries":[{"key":"dc:subject","label":"Dc Subject","values":["Applied mathematics","Mathematics"]}]},{"id":"language_rights","label":"Language and Rights","entries":[{"key":"dc:language","label":"Dc Language","values":["eng"]},{"key":"dc:rights","label":"Dc Rights","values":["In Copyright"]}]},{"id":"identifiers","label":"Identifiers","entries":[{"key":"dc:identifier","label":"Identifier","values":["etd:52577","local:Krupansky_grad.msu_0128D_21530","https://doi.org/doi:10.25335/2pa9-7w77"]}]},{"id":"additional","label":"Additional Metadata","entries":[{"key":"dc:description","label":"Description","values":["Thesis (Ph.D.)--Michigan State University. Applied Mathematics - Doctor of Philosophy, 2025","This dissertation collects results in applying neural networks to problems associated with fluid moment models in kinetic gas dynamics. Governed by the Boltzmann transport equation, gas dynamic problems are high dimensional problems and fluid moment methods are an approach to reduce the dimensionality to facilitate computation. But this reduction in dimension risks losing features and properties that are present in the higher dimension setting.We present our work developing neural network methods to solve the moment closure problem to the Boltzmann equation with the BGK approximation. Our network is trained on higher dimensional data to create closures that capture kinetic features while maintaining lower dimensional moment models. Then we turn our attention to applying neural networks to approximate Maxwellian boundary conditions for the moment problem, again with the hope of preserving kinetic features of the boundary in moment models. Finally, we explore properties of recently developed neural network architectures that allow Lipschitz and bi-Lipschitz control between layers. This is done to exposition the features of these new architectures and investigate if applying this architecture to our neural network closures improves performance and accuracy in difficult regions.","Description based on online resource. Title from PDF t.p. (Michigan State University Fedora Repository, viewed ).","Includes bibliographical references."]},{"key":"dc:format","label":"Dc Format","values":["application/pdf","1 online resource (148 p.) : ill."]},{"key":"dc:title","label":"Title","values":["Structure Preserving, Deep Learning Methods for Fluid Moment Models"]}]}],"canonical_facts":{"dc:contributor":["Christlieb, Andrew J","O'Shea, Brian W","Qian, Jianliang","Lei, Huan"],"dc:creator":["Krupansky, Nicholas Anthony"],"dc:date":["2025"],"dc:description":["Thesis (Ph.D.)--Michigan State University. Applied Mathematics - Doctor of Philosophy, 2025","This dissertation collects results in applying neural networks to problems associated with fluid moment models in kinetic gas dynamics. Governed by the Boltzmann transport equation, gas dynamic problems are high dimensional problems and fluid moment methods are an approach to reduce the dimensionality to facilitate computation. But this reduction in dimension risks losing features and properties that are present in the higher dimension setting.We present our work developing neural network methods to solve the moment closure problem to the Boltzmann equation with the BGK approximation. Our network is trained on higher dimensional data to create closures that capture kinetic features while maintaining lower dimensional moment models. Then we turn our attention to applying neural networks to approximate Maxwellian boundary conditions for the moment problem, again with the hope of preserving kinetic features of the boundary in moment models. Finally, we explore properties of recently developed neural network architectures that allow Lipschitz and bi-Lipschitz control between layers. This is done to exposition the features of these new architectures and investigate if applying this architecture to our neural network closures improves performance and accuracy in difficult regions.","Description based on online resource. Title from PDF t.p. (Michigan State University Fedora Repository, viewed ).","Includes bibliographical references."],"dc:format":["application/pdf","1 online resource (148 p.) : ill."],"dc:identifier":["etd:52577","local:Krupansky_grad.msu_0128D_21530","https://doi.org/doi:10.25335/2pa9-7w77"],"dc:language":["eng"],"dc:relation":["Electronic Theses & Dissertations"],"dc:rights":["In Copyright"],"dc:subject":["Applied mathematics","Mathematics"],"dc:title":["Structure Preserving, Deep Learning Methods for Fluid Moment Models"],"dc:type":["text","Electronic dissertations"]},"updated_at":"2026-11-18T18:59:09Z"}