{"id":{"repo_id":"mtu","oai_identifier":"oai:digitalcommons.mtu.edu:etdr-3175"},"canonical_url":"https://search.ndltd.org/etd/mtu/oai:digitalcommons.mtu.edu:etdr-3175","repository":{"repo_id":"mtu","name":"Michigan Technological University","base_url":"https://digitalcommons.mtu.edu/do/oai/"},"display":{"title":"Multilayer Adaptive and Cooperative Control of Pipeline Systems: From Local Loop Adaptation to Distributed Game-Theoretic Coordination","abstract":"This dissertation presents a unified adaptive and coordinated control architecture for multi-station centrifugal pump pipeline systems operating under dynamic hydraulic conditions. Local controller adaptability is achieved using a variable-forgetting-factor Recursive Least Squares (RLS) algorithm, which continuously estimates process parameters to adjust controller gains in real time. To prevent excessive effort and controller-induced instability, a reinforcement-learning supervisory layer based on Q-learning is implemented to determine optimal adaptation-activation policies. For pipeline systems with multiple pump stations, a game-theoretic coordination layer is introduced to balance discharge-pressure contributions, minimize pressure oscillations, and improve energy distribution across geographically distributed assets. Cooperative and non-cooperative strategies are analyzed, and Nash equilibria are shown to support graceful degradation in the presence of communication loss. Simulation results show significant reductions in steady-state error, improved transient recovery, and improved network pressure uniformity compared with conventional PI control. The unified framework provides a scalable, resilient approach to intelligent pipeline operation and lays the foundation for further research into learning-enabled, multi-agent, autonomous industrial control systems.","abstract_html":"This dissertation presents a unified adaptive and coordinated control architecture for multi-station centrifugal pump pipeline systems operating under dynamic hydraulic conditions. Local controller adaptability is achieved using a variable-forgetting-factor Recursive Least Squares (RLS) algorithm, which continuously estimates process parameters to adjust controller gains in real time. To prevent excessive effort and controller-induced instability, a reinforcement-learning supervisory layer based on Q-learning is implemented to determine optimal adaptation-activation policies. For pipeline systems with multiple pump stations, a game-theoretic coordination layer is introduced to balance discharge-pressure contributions, minimize pressure oscillations, and improve energy distribution across geographically distributed assets. Cooperative and non-cooperative strategies are analyzed, and Nash equilibria are shown to support graceful degradation in the presence of communication loss. Simulation results show significant reductions in steady-state error, improved transient recovery, and improved network pressure uniformity compared with conventional PI control. The unified framework provides a scalable, resilient approach to intelligent pipeline operation and lays the foundation for further research into learning-enabled, multi-agent, autonomous industrial control systems.","abstract_has_math":false,"creators":["Brattley, David"],"institution":null,"degree_name":"Doctor of Philosophy in Mechanical Engineering-Engineering Mechanics (PhD)","degree_level":"Open Access Dissertation","degree_discipline":"Department of Mechanical and Aerospace Engineering","degree_department":null,"school":null,"contributors":["Wayne Weaver"],"advisors":[],"committee_chairs":[],"committee_members":["Gordon Parker","Shangyan Zou","Flavio Bezerra Costa"],"year":null,"date_issued":"","date_published":null,"updated_at":"2026-11-19T08:19:26Z","subjects":[],"languages":[],"rights":[],"rights_urls":[],"identifier_entries":[{"key":"dc:identifier","label":"Identifier","values":["https://digitalcommons.mtu.edu/etdr/2098"],"render_values":[{"text":"https://digitalcommons.mtu.edu/etdr/2098","href":"https://digitalcommons.mtu.edu/etdr/2098","code":true}]}],"person_identifier_entries":[]},"links":{"outbound_url":"https://doi.org/10.37099/mtu.dc.etdr/2098","outbound_label":"DOI","outbound_source":"dc:identifier","record_url":"https://doi.org/10.37099/mtu.dc.etdr/2098","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://digitalcommons.mtu.edu/do/oai/?verb=GetRecord&metadataPrefix=oai_etdms&identifier=oai%3Adigitalcommons.mtu.edu%3Aetdr-3175","prefix":"oai_etdms"},"metadata_groups":[{"id":"people","label":"People","entries":[{"key":"dc:contributor","label":"Contributor","values":["Wayne Weaver"]},{"key":"dc:contributor.committeemember","label":"Committee Member","values":["Gordon Parker","Shangyan Zou","Flavio Bezerra Costa"]},{"key":"dc:creator","label":"Author","values":["Brattley, David"]}]},{"id":"academic_context","label":"Academic Context","entries":[{"key":"dc:date.created","label":"Dc Date Created","values":["2026-01-01T08:00:00Z"]},{"key":"thesis:degree_discipline","label":"Discipline","values":["Department of Mechanical and Aerospace Engineering"]},{"key":"thesis:degree_level","label":"Degree Level","values":["Open Access Dissertation"]},{"key":"thesis:degree_name","label":"Degree Name","values":["Doctor of Philosophy in Mechanical Engineering-Engineering Mechanics (PhD)"]}]},{"id":"identifiers","label":"Identifiers","entries":[{"key":"dc:identifier","label":"Identifier","values":["10.37099/mtu.dc.etdr/2098","https://digitalcommons.mtu.edu/etdr/2098"]}]},{"id":"additional","label":"Additional Metadata","entries":[{"key":"dc:description.abstract","label":"Abstract","values":["This dissertation presents a unified adaptive and coordinated control architecture for multi-station centrifugal pump pipeline systems operating under dynamic hydraulic conditions. Local controller adaptability is achieved using a variable-forgetting-factor Recursive Least Squares (RLS) algorithm, which continuously estimates process parameters to adjust controller gains in real time. To prevent excessive effort and controller-induced instability, a reinforcement-learning supervisory layer based on Q-learning is implemented to determine optimal adaptation-activation policies. For pipeline systems with multiple pump stations, a game-theoretic coordination layer is introduced to balance discharge-pressure contributions, minimize pressure oscillations, and improve energy distribution across geographically distributed assets. Cooperative and non-cooperative strategies are analyzed, and Nash equilibria are shown to support graceful degradation in the presence of communication loss. Simulation results show significant reductions in steady-state error, improved transient recovery, and improved network pressure uniformity compared with conventional PI control. The unified framework provides a scalable, resilient approach to intelligent pipeline operation and lays the foundation for further research into learning-enabled, multi-agent, autonomous industrial control systems."]},{"key":"dc:title","label":"Title","values":["Multilayer Adaptive and Cooperative Control of Pipeline Systems: From Local Loop Adaptation to Distributed Game-Theoretic Coordination"]}]}],"canonical_facts":{"dc:contributor":["Wayne Weaver"],"dc:contributor.committeemember":["Gordon Parker","Shangyan Zou","Flavio Bezerra Costa"],"dc:creator":["Brattley, David"],"dc:date.created":["2026-01-01T08:00:00Z"],"dc:description.abstract":["This dissertation presents a unified adaptive and coordinated control architecture for multi-station centrifugal pump pipeline systems operating under dynamic hydraulic conditions. Local controller adaptability is achieved using a variable-forgetting-factor Recursive Least Squares (RLS) algorithm, which continuously estimates process parameters to adjust controller gains in real time. To prevent excessive effort and controller-induced instability, a reinforcement-learning supervisory layer based on Q-learning is implemented to determine optimal adaptation-activation policies. For pipeline systems with multiple pump stations, a game-theoretic coordination layer is introduced to balance discharge-pressure contributions, minimize pressure oscillations, and improve energy distribution across geographically distributed assets. Cooperative and non-cooperative strategies are analyzed, and Nash equilibria are shown to support graceful degradation in the presence of communication loss. Simulation results show significant reductions in steady-state error, improved transient recovery, and improved network pressure uniformity compared with conventional PI control. The unified framework provides a scalable, resilient approach to intelligent pipeline operation and lays the foundation for further research into learning-enabled, multi-agent, autonomous industrial control systems."],"dc:identifier":["10.37099/mtu.dc.etdr/2098","https://digitalcommons.mtu.edu/etdr/2098"],"dc:title":["Multilayer Adaptive and Cooperative Control of Pipeline Systems: From Local Loop Adaptation to Distributed Game-Theoretic Coordination"],"thesis:degree_discipline":["Department of Mechanical and Aerospace Engineering"],"thesis:degree_level":["Open Access Dissertation"],"thesis:degree_name":["Doctor of Philosophy in Mechanical Engineering-Engineering Mechanics (PhD)"]},"updated_at":"2026-11-19T08:19:26Z"}