{"id":{"repo_id":"cambridge","oai_identifier":"oai:www.repository.cam.ac.uk:1810/358328"},"canonical_url":"https://search.dev.ndltd.org/etd/cambridge/oai:www.repository.cam.ac.uk:1810/358328","repository":{"repo_id":"cambridge","name":"Cambridge University","base_url":"https://api.repository.cam.ac.uk/server/oai/request"},"display":{"title":"Transfer Learning for Accelerated Process Development","abstract":"One of the greatest challenges in process development is the limited amount of data that can be collected. Techniques that can draw insights from this limited data have the possibility to accelerate process development. This thesis presents a collection of studies on using transfer learning to accelerate various aspects of process development. Part I focuses on reaction optimization, where I propose a benchmarking framework for comparing machine learning strategies for reaction optimization and demonstrate the benefits of using multi-task learning to accelerate chemical reaction optimization. In Part II, I explore the use of reinforcement learning and multi-fidelity Bayesian optimization for accelerating feedback controller tuning, specifically for distillation control systems. Finally, in Part III, I take two perspectives on using machine learning for predictive thermodynamics, a key aspect of process modelling. I introduce *DeepGamma* for predicting activity coefficients and ML-SAFT for predicting PCP-SAFT parameters, showing steps towards improving thermodynamic predictions using transfer learning. Together, all of these studies demonstrate the potential of transfer learning to accelerate process development, providing valuable insights for future research and practical applications.","abstract_html":"One of the greatest challenges in process development is the limited amount of data that can be collected. Techniques that can draw insights from this limited data have the possibility to accelerate process development. This thesis presents a collection of studies on using transfer learning to accelerate various aspects of process development. Part I focuses on reaction optimization, where I propose a benchmarking framework for comparing machine learning strategies for reaction optimization and demonstrate the benefits of using multi-task learning to accelerate chemical reaction optimization. In Part II, I explore the use of reinforcement learning and multi-fidelity Bayesian optimization for accelerating feedback controller tuning, specifically for distillation control systems. Finally, in Part III, I take two perspectives on using machine learning for predictive thermodynamics, a key aspect of process modelling. I introduce *DeepGamma* for predicting activity coefficients and ML-SAFT for predicting PCP-SAFT parameters, showing steps towards improving thermodynamic predictions using transfer learning. Together, all of these studies demonstrate the potential of transfer learning to accelerate process development, providing valuable insights for future research and practical applications.","abstract_has_math":false,"creators":["Felton, Kobi"],"institution":"University of Cambridge","degree_name":"Doctor of Philosophy (PhD)","degree_level":"Doctoral","degree_discipline":null,"degree_department":null,"school":null,"contributors":[],"advisors":["Lapkin, Alexei"],"committee_chairs":[],"committee_members":[],"year":2023,"date_issued":"2023-07-27","date_published":"2023-07-27","updated_at":"2026-07-22T22:24:18Z","subjects":["Bayesian optimization","deep learning","machine learning","PC-SAFT","thermodynamics","transfer learning"],"languages":["eng"],"rights":[],"rights_urls":["https://apollo8-f-pro.lib.cam.ac.uk/bitstreams/e5722236-175f-42de-9461-f846774a2f57/download","https://www.rioxx.net/licenses/all-rights-reserved/"],"identifier_entries":[{"key":"dc:creator.authoridentifier","label":"Author Identifier","values":["0000000236164766"],"render_values":[{"text":"0000-0002-3616-4766","href":"https://orcid.org/0000-0002-3616-4766","code":true}]}]},"links":{"outbound_url":"https://doi.org/10.17863/CAM.102031","outbound_label":"DOI","outbound_source":"dc:identifier.doi"},"metadata_groups":[{"id":"people","label":"People","entries":[{"key":"dc:contributor.advisor","label":"Advisor","values":["Lapkin, Alexei"]},{"key":"dc:contributor.sponsor","label":"Sponsor","values":["Marshall Scholarship and BASF SE."]},{"key":"dc:creator","label":"Author","values":["Felton, Kobi"]},{"key":"dc:creator.authoridentifier","label":"Author Identifier","values":["0000000236164766"]}]},{"id":"academic_context","label":"Academic Context","entries":[{"key":"dc:date.issued","label":"Date","values":["2023-07-27"]},{"key":"dc:publisher.institution","label":"Dc Publisher Institution","values":["University of Cambridge"]},{"key":"dc:relation.isreferencedby.uri","label":"Dc Relation Isreferencedby URI","values":["https://www.repository.cam.ac.uk/handle/1810/358328"]},{"key":"dc:type","label":"Dc Type","values":["Thesis"]},{"key":"dc:type.qualificationlevel","label":"Dc Type Qualificationlevel","values":["Doctoral"]},{"key":"dc:type.qualificationname","label":"Dc Type Qualificationname","values":["Doctor of Philosophy (PhD)"]}]},{"id":"subjects_keywords","label":"Subjects and Keywords","entries":[{"key":"dc:subject","label":"Dc Subject","values":["Bayesian optimization","deep learning","machine learning","PC-SAFT","thermodynamics","transfer learning"]}]},{"id":"language_rights","label":"Language and Rights","entries":[{"key":"dc:language","label":"Dc Language","values":["eng"]},{"key":"dc:rights","label":"Dc Rights","values":["https://apollo8-f-pro.lib.cam.ac.uk/bitstreams/e5722236-175f-42de-9461-f846774a2f57/download","https://www.rioxx.net/licenses/all-rights-reserved/"]}]},{"id":"identifiers","label":"Identifiers","entries":[{"key":"dc:identifier.doi","label":"DOI","values":["https://doi.org/10.17863/CAM.102031"]},{"key":"dc:identifier.uri","label":"Identifier URI","values":["https://apollo8-f-pro.lib.cam.ac.uk/bitstreams/49783d3e-a6d0-401d-a54a-7fbda9c9aa89/download"]}]},{"id":"additional","label":"Additional Metadata","entries":[{"key":"dc:description.abstract","label":"Abstract","values":["One of the greatest challenges in process development is the limited amount of data that can be collected. Techniques that can draw insights from this limited data have the possibility to accelerate process development. This thesis presents a collection of studies on using transfer learning to accelerate various aspects of process development. Part I focuses on reaction optimization, where I propose a benchmarking framework for comparing machine learning strategies for reaction optimization and demonstrate the benefits of using multi-task learning to accelerate chemical reaction optimization. In Part II, I explore the use of reinforcement learning and multi-fidelity Bayesian optimization for accelerating feedback controller tuning, specifically for distillation control systems. Finally, in Part III, I take two perspectives on using machine learning for predictive thermodynamics, a key aspect of process modelling. I introduce *DeepGamma* for predicting activity coefficients and ML-SAFT for predicting PCP-SAFT parameters, showing steps towards improving thermodynamic predictions using transfer learning. Together, all of these studies demonstrate the potential of transfer learning to accelerate process development, providing valuable insights for future research and practical applications."]},{"key":"dc:format.checksum.md5","label":"Dc Format Checksum Md5","values":["10fdd268ee9f303a43724d0f62d57850","87eda9de84448d1f82354d60eee3eb5f"]},{"key":"dc:title","label":"Title","values":["Transfer Learning for Accelerated Process Development"]}]}],"canonical_facts":{"dc:contributor.advisor":["Lapkin, Alexei"],"dc:contributor.sponsor":["Marshall Scholarship and BASF SE."],"dc:creator":["Felton, Kobi"],"dc:creator.authoridentifier":["0000000236164766"],"dc:date.issued":["2023-07-27"],"dc:description.abstract":["One of the greatest challenges in process development is the limited amount of data that can be collected. Techniques that can draw insights from this limited data have the possibility to accelerate process development. This thesis presents a collection of studies on using transfer learning to accelerate various aspects of process development. Part I focuses on reaction optimization, where I propose a benchmarking framework for comparing machine learning strategies for reaction optimization and demonstrate the benefits of using multi-task learning to accelerate chemical reaction optimization. In Part II, I explore the use of reinforcement learning and multi-fidelity Bayesian optimization for accelerating feedback controller tuning, specifically for distillation control systems. Finally, in Part III, I take two perspectives on using machine learning for predictive thermodynamics, a key aspect of process modelling. I introduce *DeepGamma* for predicting activity coefficients and ML-SAFT for predicting PCP-SAFT parameters, showing steps towards improving thermodynamic predictions using transfer learning. Together, all of these studies demonstrate the potential of transfer learning to accelerate process development, providing valuable insights for future research and practical applications."],"dc:format.checksum.md5":["10fdd268ee9f303a43724d0f62d57850","87eda9de84448d1f82354d60eee3eb5f"],"dc:identifier.doi":["https://doi.org/10.17863/CAM.102031"],"dc:identifier.uri":["https://apollo8-f-pro.lib.cam.ac.uk/bitstreams/49783d3e-a6d0-401d-a54a-7fbda9c9aa89/download"],"dc:language":["eng"],"dc:publisher.institution":["University of Cambridge"],"dc:relation.isreferencedby.uri":["https://www.repository.cam.ac.uk/handle/1810/358328"],"dc:rights":["https://apollo8-f-pro.lib.cam.ac.uk/bitstreams/e5722236-175f-42de-9461-f846774a2f57/download","https://www.rioxx.net/licenses/all-rights-reserved/"],"dc:subject":["Bayesian optimization","deep learning","machine learning","PC-SAFT","thermodynamics","transfer learning"],"dc:title":["Transfer Learning for Accelerated Process Development"],"dc:type":["Thesis"],"dc:type.qualificationlevel":["Doctoral"],"dc:type.qualificationname":["Doctor of Philosophy (PhD)"]},"updated_at":"2026-07-22T22:24:18Z"}