{"id":{"repo_id":"mcmaster","oai_identifier":"oai:macsphere.mcmaster.ca:11375/33312"},"canonical_url":"https://search.ndltd.org/etd/mcmaster/oai:macsphere.mcmaster.ca:11375/33312","repository":{"repo_id":"mcmaster","name":"McMaster University","base_url":"https://macsphere.mcmaster.ca/server/oai/request"},"display":{"title":"A Computational Framework For Two-Phase Decomposition In Plant Modeling, Optimization, And Stochastic Analysis","abstract":"Process plants are often represented by several models built for different decision layers, including rigorous simulation, operating-condition optimization, production scheduling, and long-horizon planning. When these models are developed independently, they may use different abstractions of the same plant and can produce inconsistent solutions. This thesis develops a computational framework for coordinating plant models across levels of detail while preserving enough accuracy for plant-wide optimization and related decision problems. The central contribution is a two-phase decomposition algorithm based on a unified process-network representation. The method reformulates plant calculations around component mass flows and local thermophysical-property approximations, then partitions the model into linear mass and approximate energy-balance equations and higher-accuracy energy-balance calculations. Phase I solves the plant-wide linear model, Phase II updates thermophysical and unit-model parameters at the computed flows, and the process is repeated until the abstraction levels are consistent. The approach is validated on a blue hydrogen plant case study, where the hybrid model gives solutions within 1\\% of a rigorous plant model and the computational effort grows approximately linearly with the number of time periods. The thesis also develops a configurable software framework that represents process units, streams, variables, equations, and solution workflows needed to automate the two-phase methodology. Finally, the framework is extended to a focused two-stage stochastic optimization problem for CHP/turbine sizing under electricity-price and hydrogen-demand uncertainty. In that extension, the stochastic formulation selects a smaller turbine capacity than the deterministic benchmark and demonstrates how the deterministic methodology can support uncertainty-aware process optimization.","abstract_html":"Process plants are often represented by several models built for different decision layers, including rigorous simulation, operating-condition optimization, production scheduling, and long-horizon planning. When these models are developed independently, they may use different abstractions of the same plant and can produce inconsistent solutions. This thesis develops a computational framework for coordinating plant models across levels of detail while preserving enough accuracy for plant-wide optimization and related decision problems. The central contribution is a two-phase decomposition algorithm based on a unified process-network representation. The method reformulates plant calculations around component mass flows and local thermophysical-property approximations, then partitions the model into linear mass and approximate energy-balance equations and higher-accuracy energy-balance calculations. Phase I solves the plant-wide linear model, Phase II updates thermophysical and unit-model parameters at the computed flows, and the process is repeated until the abstraction levels are consistent. The approach is validated on a blue hydrogen plant case study, where the hybrid model gives solutions within 1\\% of a rigorous plant model and the computational effort grows approximately linearly with the number of time periods. The thesis also develops a configurable software framework that represents process units, streams, variables, equations, and solution workflows needed to automate the two-phase methodology. Finally, the framework is extended to a focused two-stage stochastic optimization problem for CHP/turbine sizing under electricity-price and hydrogen-demand uncertainty. In that extension, the stochastic formulation selects a smaller turbine capacity than the deterministic benchmark and demonstrates how the deterministic methodology can support uncertainty-aware process optimization.","abstract_has_math":false,"creators":["Maghsoudi, Farbod"],"institution":null,"degree_name":null,"degree_level":null,"degree_discipline":null,"degree_department":"Chemical Engineering","school":null,"contributors":[],"advisors":["Mahalec, Vladimir"],"committee_chairs":[],"committee_members":[],"year":2026,"date_issued":"2026","date_published":"2026","updated_at":"2026-11-18T19:32:37Z","subjects":["plant-wide optimization","process modeling","two-phase decomposition","hybrid plant models","thermophysical-property approximation","blue hydrogen","Pyomo","stochastic optimization"],"languages":["en"],"rights":[],"rights_urls":[],"identifier_entries":[{"key":"dc:identifier.uri","label":"Identifier URI","values":["https://doi.org/10.71548/174"],"render_values":[{"text":"https://doi.org/10.71548/174","href":"https://doi.org/10.71548/174","code":true}]}],"person_identifier_entries":[]},"links":{"outbound_url":"https://hdl.handle.net/11375/33312","outbound_label":"Handle","outbound_source":"dc:identifier.uri","record_url":"https://hdl.handle.net/11375/33312","record_label":"Repository record","record_kind":"handle","record_source":"dc:identifier.uri","download_url":null,"download_label":null,"download_source":null,"access_status":null,"embargo_until":null,"access_note":null},"source_record":{"url":"https://macsphere.mcmaster.ca/server/oai/request?verb=GetRecord&metadataPrefix=dim&identifier=oai%3Amacsphere.mcmaster.ca%3A11375%2F33312","prefix":"dim"},"metadata_groups":[{"id":"people","label":"People","entries":[{"key":"dc:contributor.advisor","label":"Advisor","values":["Mahalec, Vladimir"]},{"key":"dc:contributor.department","label":"Department","values":["Chemical Engineering"]},{"key":"dc:creator","label":"Author","values":["Maghsoudi, Farbod"]}]},{"id":"academic_context","label":"Academic Context","entries":[{"key":"dc:date.accessioned","label":"Dc Date Accessioned","values":["2026-06-30T19:33:09Z"]},{"key":"dc:date.issued","label":"Date","values":["2026"]},{"key":"dc:type","label":"Dc Type","values":["Thesis"]}]},{"id":"subjects_keywords","label":"Subjects and Keywords","entries":[{"key":"dc:subject","label":"Dc Subject","values":["plant-wide optimization","process modeling","two-phase decomposition","hybrid plant models","thermophysical-property approximation","blue hydrogen","Pyomo","stochastic optimization"]}]},{"id":"language_rights","label":"Language and Rights","entries":[{"key":"dc:language.iso","label":"Language (ISO)","values":["en"]}]},{"id":"identifiers","label":"Identifiers","entries":[{"key":"dc:identifier.uri","label":"Identifier URI","values":["https://hdl.handle.net/11375/33312","https://doi.org/10.71548/174"]}]},{"id":"additional","label":"Additional Metadata","entries":[{"key":"dc:description.abstract","label":"Abstract","values":["Process plants are often represented by several models built for different decision layers, including rigorous simulation, operating-condition optimization, production scheduling, and long-horizon planning. When these models are developed independently, they may use different abstractions of the same plant and can produce inconsistent solutions. This thesis develops a computational framework for coordinating plant models across levels of detail while preserving enough accuracy for plant-wide optimization and related decision problems. The central contribution is a two-phase decomposition algorithm based on a unified process-network representation. The method reformulates plant calculations around component mass flows and local thermophysical-property approximations, then partitions the model into linear mass and approximate energy-balance equations and higher-accuracy energy-balance calculations. Phase I solves the plant-wide linear model, Phase II updates thermophysical and unit-model parameters at the computed flows, and the process is repeated until the abstraction levels are consistent. The approach is validated on a blue hydrogen plant case study, where the hybrid model gives solutions within 1\\% of a rigorous plant model and the computational effort grows approximately linearly with the number of time periods. The thesis also develops a configurable software framework that represents process units, streams, variables, equations, and solution workflows needed to automate the two-phase methodology. Finally, the framework is extended to a focused two-stage stochastic optimization problem for CHP/turbine sizing under electricity-price and hydrogen-demand uncertainty. In that extension, the stochastic formulation selects a smaller turbine capacity than the deterministic benchmark and demonstrates how the deterministic methodology can support uncertainty-aware process optimization."]},{"key":"dc:description.degree","label":"Dc Description Degree","values":["Master of Applied Science (MASc)"]},{"key":"dc:description.sponsorship","label":"Dc Description Sponsorship","values":["Natural Sciences and Engineering Research Council of Canada (NSERC); McMaster Advanced Control Consortium (MACC)"]},{"key":"dc:title","label":"Title","values":["A Computational Framework For Two-Phase Decomposition In Plant Modeling, Optimization, And Stochastic Analysis"]},{"key":"others:access_status","label":"Others Access Status","values":["embargo"]}]}],"canonical_facts":{"dc:contributor.advisor":["Mahalec, Vladimir"],"dc:contributor.department":["Chemical Engineering"],"dc:creator":["Maghsoudi, Farbod"],"dc:date.accessioned":["2026-06-30T19:33:09Z"],"dc:date.issued":["2026"],"dc:description.abstract":["Process plants are often represented by several models built for different decision layers, including rigorous simulation, operating-condition optimization, production scheduling, and long-horizon planning. When these models are developed independently, they may use different abstractions of the same plant and can produce inconsistent solutions. This thesis develops a computational framework for coordinating plant models across levels of detail while preserving enough accuracy for plant-wide optimization and related decision problems. The central contribution is a two-phase decomposition algorithm based on a unified process-network representation. The method reformulates plant calculations around component mass flows and local thermophysical-property approximations, then partitions the model into linear mass and approximate energy-balance equations and higher-accuracy energy-balance calculations. Phase I solves the plant-wide linear model, Phase II updates thermophysical and unit-model parameters at the computed flows, and the process is repeated until the abstraction levels are consistent. The approach is validated on a blue hydrogen plant case study, where the hybrid model gives solutions within 1\\% of a rigorous plant model and the computational effort grows approximately linearly with the number of time periods. The thesis also develops a configurable software framework that represents process units, streams, variables, equations, and solution workflows needed to automate the two-phase methodology. Finally, the framework is extended to a focused two-stage stochastic optimization problem for CHP/turbine sizing under electricity-price and hydrogen-demand uncertainty. In that extension, the stochastic formulation selects a smaller turbine capacity than the deterministic benchmark and demonstrates how the deterministic methodology can support uncertainty-aware process optimization."],"dc:description.degree":["Master of Applied Science (MASc)"],"dc:description.sponsorship":["Natural Sciences and Engineering Research Council of Canada (NSERC); McMaster Advanced Control Consortium (MACC)"],"dc:identifier.uri":["https://hdl.handle.net/11375/33312","https://doi.org/10.71548/174"],"dc:language.iso":["en"],"dc:subject":["plant-wide optimization","process modeling","two-phase decomposition","hybrid plant models","thermophysical-property approximation","blue hydrogen","Pyomo","stochastic optimization"],"dc:title":["A Computational Framework For Two-Phase Decomposition In Plant Modeling, Optimization, And Stochastic Analysis"],"dc:type":["Thesis"],"others:access_status":["embargo"]},"updated_at":"2026-11-18T19:32:37Z"}