{"id":{"repo_id":"birmingham","oai_identifier":"oai:etheses.bham.ac.uk:12003"},"canonical_url":"https://search.ndltd.org/etd/birmingham/oai:etheses.bham.ac.uk:12003","repository":{"repo_id":"birmingham","name":"University of Birmingham","base_url":"https://etheses.bham.ac.uk/cgi/oai2"},"display":{"title":"Artificial intelligence and chemical kinetics enabled property-oriented fuel design for internal combustion engine","abstract":"Fuel Genome Project aims at addressing the forward problem of fuel property prediction and the inverse problems of molecule design, retrosynthesis and reaction condition prediction. This work primarily addresses the forward problem by integrating feature engineering theory, artificial intelligence (AI) technologies, gas-phase chemical kinetics. Group contribution method (GCM) is utilized to establish the GCM-UOB (University of Birmingham) 1.0 system with 22 molecular descriptors and the surrogate formulation is to minimize the difference of functional group fragments between target fuel and surrogate. The improved QSPR (quantitative structure–activity relationship)-UOB 2.0 system with 32 molecular features couples with machine learning (ML) algorithms to establish the regression models for fuel ignition quality prediction. QSPR-UOB 3.0 scheme expands to 42 molecular descriptors to improve the molecular resolution of aromatics and specific fuel types. The obtained structural features combining with ML algorithms enable to predict 15 physicochemical properties with high fidelity and efficiency. In addition to the technical route of ML-QSPR models, another route of deep learning-convolution neural network (DL-CNN) is proposed for property prediction and yield sooting index (YSI) is taken as a case study. The predicted accuracy of DL-CNN is inferior to the ML-QSPR model at its current status, but its benefit of automated feature extraction and rapid advance in classification problem make it a promising solution for regression problem. A high-throughput fuel screening is performed to identify the molecules with desired properties for both spark ignition (SI) and compression ignition (CI) engines which contains the Tier 1 physicochemical properties screening (based on the ML-QSPR models) and Tier 2 chemical kinetic screening (based on the detailed chemical mechanisms). Polyoxymethylene dimethyl ether 3 (PODE3) and diethoxymethane (DEM) are promising carbon-neutral fuels for CI engines and they are recommended by the virtual screening results. Their ignition delay time, laminar flame speed and dominant reactions of PODE3 and DEM are examined by chemical kinetics and a new DEM mechanism including both low and high-temperature reactions is constructed. Concluding remarks and research prospects are summarized in the final section.","abstract_html":"Fuel Genome Project aims at addressing the forward problem of fuel property prediction and the inverse problems of molecule design, retrosynthesis and reaction condition prediction. This work primarily addresses the forward problem by integrating feature engineering theory, artificial intelligence (AI) technologies, gas-phase chemical kinetics. Group contribution method (GCM) is utilized to establish the GCM-UOB (University of Birmingham) 1.0 system with 22 molecular descriptors and the surrogate formulation is to minimize the difference of functional group fragments between target fuel and surrogate. The improved QSPR (quantitative structure–activity relationship)-UOB 2.0 system with 32 molecular features couples with machine learning (ML) algorithms to establish the regression models for fuel ignition quality prediction. QSPR-UOB 3.0 scheme expands to 42 molecular descriptors to improve the molecular resolution of aromatics and specific fuel types. The obtained structural features combining with ML algorithms enable to predict 15 physicochemical properties with high fidelity and efficiency. In addition to the technical route of ML-QSPR models, another route of deep learning-convolution neural network (DL-CNN) is proposed for property prediction and yield sooting index (YSI) is taken as a case study. The predicted accuracy of DL-CNN is inferior to the ML-QSPR model at its current status, but its benefit of automated feature extraction and rapid advance in classification problem make it a promising solution for regression problem. A high-throughput fuel screening is performed to identify the molecules with desired properties for both spark ignition (SI) and compression ignition (CI) engines which contains the Tier 1 physicochemical properties screening (based on the ML-QSPR models) and Tier 2 chemical kinetic screening (based on the detailed chemical mechanisms). Polyoxymethylene dimethyl ether 3 (PODE3) and diethoxymethane (DEM) are promising carbon-neutral fuels for CI engines and they are recommended by the virtual screening results. Their ignition delay time, laminar flame speed and dominant reactions of PODE3 and DEM are examined by chemical kinetics and a new DEM mechanism including both low and high-temperature reactions is constructed. Concluding remarks and research prospects are summarized in the final section.","abstract_has_math":false,"creators":["Li, Runzhao"],"institution":"University of Birmingham","degree_name":"d_ph","degree_level":"d_ph","degree_discipline":null,"degree_department":null,"school":null,"contributors":[],"advisors":[],"committee_chairs":[],"committee_members":[],"year":2021,"date_issued":"2021-12","date_published":"2021-12","updated_at":"2026-11-19T09:31:36Z","subjects":["QA Mathematics","QA75 Electronic computers. Computer science","TJ Mechanical engineering and machinery","TP Chemical technology"],"languages":["eng"],"rights":[],"rights_urls":[],"identifier_entries":[],"person_identifier_entries":[{"key":"person:identifier.orcid","label":"ORCID","values":["0000000151209849"],"render_values":[{"text":"0000-0001-5120-9849","href":"https://orcid.org/0000-0001-5120-9849","code":true}]}]},"links":{"outbound_url":null,"outbound_label":null,"outbound_source":null,"record_url":null,"record_label":null,"record_kind":null,"record_source":null,"download_url":"http://etheses.bham.ac.uk//id/eprint/12003/7/Li2021PhD_Redacted.pdf","download_label":"Download PDF","download_source":"dc:identifier.uri","access_status":null,"embargo_until":null,"access_note":null},"source_record":{"url":"https://etheses.bham.ac.uk/cgi/oai2?verb=GetRecord&metadataPrefix=uketd_dc&identifier=oai%3Aetheses.bham.ac.uk%3A12003","prefix":"uketd_dc"},"metadata_groups":[{"id":"people","label":"People","entries":[{"key":"dc:contributor.sponsor","label":"Sponsor","values":["epsrc","other"]},{"key":"dc:creator","label":"Author","values":["Li, Runzhao"]},{"key":"person:identifier.orcid","label":"ORCID","values":["0000000151209849"]}]},{"id":"academic_context","label":"Academic Context","entries":[{"key":"dc:date","label":"Dc Date","values":["2021-12-06"]},{"key":"dc:date.issued","label":"Date","values":["2021-12"]},{"key":"dc:publisher.department","label":"Dc Publisher Department","values":["College of Engineering & Physical Sciences","School of Engineering, Department of Mechanical Engineering"]},{"key":"dc:publisher.institution","label":"Dc Publisher Institution","values":["University of Birmingham"]},{"key":"dc:relation.isreferencedby","label":"Dc Relation Isreferencedby","values":["http://etheses.bham.ac.uk//id/eprint/12003/"]},{"key":"dc:type","label":"Dc Type","values":["Thesis"]},{"key":"dc:type.qualificationlevel","label":"Dc Type Qualificationlevel","values":["d_ph"]},{"key":"dc:type.qualificationname","label":"Dc Type Qualificationname","values":["d_ph"]}]},{"id":"subjects_keywords","label":"Subjects and Keywords","entries":[{"key":"dc:subject","label":"Dc Subject","values":["QA Mathematics","QA75 Electronic computers. Computer science","TJ Mechanical engineering and machinery","TP Chemical technology"]}]},{"id":"language_rights","label":"Language and Rights","entries":[{"key":"dc:language.dcterms_iso639_2","label":"Language (ISO 639-2)","values":["eng"]}]},{"id":"identifiers","label":"Identifiers","entries":[{"key":"dc:identifier.uri","label":"Identifier URI","values":["http://etheses.bham.ac.uk//id/eprint/12003/7/Li2021PhD_Redacted.pdf"]}]},{"id":"additional","label":"Additional Metadata","entries":[{"key":"dc:description.abstract","label":"Abstract","values":["Fuel Genome Project aims at addressing the forward problem of fuel property prediction and the inverse problems of molecule design, retrosynthesis and reaction condition prediction. This work primarily addresses the forward problem by integrating feature engineering theory, artificial intelligence (AI) technologies, gas-phase chemical kinetics. Group contribution method (GCM) is utilized to establish the GCM-UOB (University of Birmingham) 1.0 system with 22 molecular descriptors and the surrogate formulation is to minimize the difference of functional group fragments between target fuel and surrogate. The improved QSPR (quantitative structure–activity relationship)-UOB 2.0 system with 32 molecular features couples with machine learning (ML) algorithms to establish the regression models for fuel ignition quality prediction. QSPR-UOB 3.0 scheme expands to 42 molecular descriptors to improve the molecular resolution of aromatics and specific fuel types. The obtained structural features combining with ML algorithms enable to predict 15 physicochemical properties with high fidelity and efficiency. In addition to the technical route of ML-QSPR models, another route of deep learning-convolution neural network (DL-CNN) is proposed for property prediction and yield sooting index (YSI) is taken as a case study. The predicted accuracy of DL-CNN is inferior to the ML-QSPR model at its current status, but its benefit of automated feature extraction and rapid advance in classification problem make it a promising solution for regression problem. A high-throughput fuel screening is performed to identify the molecules with desired properties for both spark ignition (SI) and compression ignition (CI) engines which contains the Tier 1 physicochemical properties screening (based on the ML-QSPR models) and Tier 2 chemical kinetic screening (based on the detailed chemical mechanisms). Polyoxymethylene dimethyl ether 3 (PODE3) and diethoxymethane (DEM) are promising carbon-neutral fuels for CI engines and they are recommended by the virtual screening results. Their ignition delay time, laminar flame speed and dominant reactions of PODE3 and DEM are examined by chemical kinetics and a new DEM mechanism including both low and high-temperature reactions is constructed. Concluding remarks and research prospects are summarized in the final section."]},{"key":"dc:format","label":"Dc Format","values":["text"]},{"key":"dcterms:accessrights","label":"Dcterms Accessrights","values":["public"]},{"key":"dc:title","label":"Title","values":["Artificial intelligence and chemical kinetics enabled property-oriented fuel design for internal combustion engine"]}]}],"canonical_facts":{"dc:contributor.sponsor":["epsrc","other"],"dc:creator":["Li, Runzhao"],"dc:date":["2021-12-06"],"dc:date.issued":["2021-12"],"dc:description.abstract":["Fuel Genome Project aims at addressing the forward problem of fuel property prediction and the inverse problems of molecule design, retrosynthesis and reaction condition prediction. This work primarily addresses the forward problem by integrating feature engineering theory, artificial intelligence (AI) technologies, gas-phase chemical kinetics. Group contribution method (GCM) is utilized to establish the GCM-UOB (University of Birmingham) 1.0 system with 22 molecular descriptors and the surrogate formulation is to minimize the difference of functional group fragments between target fuel and surrogate. The improved QSPR (quantitative structure–activity relationship)-UOB 2.0 system with 32 molecular features couples with machine learning (ML) algorithms to establish the regression models for fuel ignition quality prediction. QSPR-UOB 3.0 scheme expands to 42 molecular descriptors to improve the molecular resolution of aromatics and specific fuel types. The obtained structural features combining with ML algorithms enable to predict 15 physicochemical properties with high fidelity and efficiency. In addition to the technical route of ML-QSPR models, another route of deep learning-convolution neural network (DL-CNN) is proposed for property prediction and yield sooting index (YSI) is taken as a case study. The predicted accuracy of DL-CNN is inferior to the ML-QSPR model at its current status, but its benefit of automated feature extraction and rapid advance in classification problem make it a promising solution for regression problem. A high-throughput fuel screening is performed to identify the molecules with desired properties for both spark ignition (SI) and compression ignition (CI) engines which contains the Tier 1 physicochemical properties screening (based on the ML-QSPR models) and Tier 2 chemical kinetic screening (based on the detailed chemical mechanisms). Polyoxymethylene dimethyl ether 3 (PODE3) and diethoxymethane (DEM) are promising carbon-neutral fuels for CI engines and they are recommended by the virtual screening results. Their ignition delay time, laminar flame speed and dominant reactions of PODE3 and DEM are examined by chemical kinetics and a new DEM mechanism including both low and high-temperature reactions is constructed. Concluding remarks and research prospects are summarized in the final section."],"dc:format":["text"],"dc:identifier.uri":["http://etheses.bham.ac.uk//id/eprint/12003/7/Li2021PhD_Redacted.pdf"],"dc:language.dcterms_iso639_2":["eng"],"dc:publisher.department":["College of Engineering & Physical Sciences","School of Engineering, Department of Mechanical Engineering"],"dc:publisher.institution":["University of Birmingham"],"dc:relation.isreferencedby":["http://etheses.bham.ac.uk//id/eprint/12003/"],"dc:subject":["QA Mathematics","QA75 Electronic computers. Computer science","TJ Mechanical engineering and machinery","TP Chemical technology"],"dcterms:accessrights":["public"],"dc:title":["Artificial intelligence and chemical kinetics enabled property-oriented fuel design for internal combustion engine"],"dc:type":["Thesis"],"dc:type.qualificationlevel":["d_ph"],"dc:type.qualificationname":["d_ph"],"person:identifier.orcid":["0000000151209849"]},"updated_at":"2026-11-19T09:31:36Z"}