Loading…
Support vector machine based emissions modeling using particle swarm optimization for homogeneous charge compression ignition engine
The internal combustion engine faces increasing societal and governmental pressure to improve both efficiency and engine out emissions. Currently, research has moved from traditional combustion methods to new highly efficient combustion strategies such as Homogeneous Charge Compression Ignition (HCC...
Saved in:
Published in: | International journal of engine research 2023-02, Vol.24 (2), p.536-551 |
---|---|
Main Authors: | , , , , , , |
Format: | Article |
Language: | English |
Subjects: | |
Citations: | Items that this one cites Items that cite this one |
Online Access: | Get full text |
Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
cited_by | cdi_FETCH-LOGICAL-c466t-2e8890a924e4204460aa34a57315ea7dbadfaaef980f350449a59e17fcae05563 |
---|---|
cites | cdi_FETCH-LOGICAL-c466t-2e8890a924e4204460aa34a57315ea7dbadfaaef980f350449a59e17fcae05563 |
container_end_page | 551 |
container_issue | 2 |
container_start_page | 536 |
container_title | International journal of engine research |
container_volume | 24 |
creator | Gordon, David Norouzi, Armin Blomeyer, Gero Bedei, Julian Aliramezani, Masoud Andert, Jakob Koch, Charles R |
description | The internal combustion engine faces increasing societal and governmental pressure to improve both efficiency and engine out emissions. Currently, research has moved from traditional combustion methods to new highly efficient combustion strategies such as Homogeneous Charge Compression Ignition (HCCI). However, predicting the exact value of engine out emissions using conventional physics-based or data-driven models is still a challenge for engine researchers due to the complexity the of combustion and emission formation. Research has focused on using Artificial Neural Networks (ANN) for this problem but ANN’s require large training datasets for acceptable accuracy. This work addresses this problem by presenting the development of a simple model for predicting the steady-state emissions of a single cylinder HCCI engine which is created using an metaheuristic optimization based Support Vector Machine (SVM). The selection of input variables to the SVM model is explored using five different feature sets, considering up to seven engine inputs. The best results are achieved with a model combining linear and squared inputs as well as cross correlations and their squares totaling 26 features. In this case the model fit represented by R2 values were between 0.72 and 0.95. The best model fits were achieved for CO and CO2, while HC and NOx models have reduced model performance. Linear and non-linear SVM models were then compared to an ANN model. This comparison showed that SVM based models were more robust to changes in feature selection and better able to avoid local minimums compared to the ANN models leading to a more consistent model prediction when limited training data is available. The proposed machine learning based HCCI emission models and the feature selection approach provide insight into optimizing the model accuracy while minimizing the computational costs. |
doi_str_mv | 10.1177/14680874211055546 |
format | article |
fullrecord | <record><control><sourceid>proquest_pubme</sourceid><recordid>TN_cdi_pubmedcentral_primary_oai_pubmedcentral_nih_gov_9902987</recordid><sourceformat>XML</sourceformat><sourcesystem>PC</sourcesystem><sage_id>10.1177_14680874211055546</sage_id><sourcerecordid>2771309981</sourcerecordid><originalsourceid>FETCH-LOGICAL-c466t-2e8890a924e4204460aa34a57315ea7dbadfaaef980f350449a59e17fcae05563</originalsourceid><addsrcrecordid>eNp1kUFv1DAQhS0EotvCD-CCLHHhkmIndhxfkFAFBalSD4WzNetMsq7iONhJEZz54Ti7pVBQL_Zhvnkzbx4hLzg75VypN1zUDWuUKDlnUkpRPyKbkgleVFzox2Sz1osVOCLHKV0zxqRQ6ik5qmqlasH1hvy8WqYpxJneoJ1DpB7szo1It5CwpehdSi6MifrQ4uDGni5pfSeIs7MD0vQNoqdhmp13P2DOLO2yzC740OOIYUnU7iD2SG3wU8S9HHX96PYsjn2e9ow86WBI-Pz2PyFfPrz_fPaxuLg8_3T27qKwoq7nosSm0Qx0KVBkm6JmAJUAqSouEVS7hbYDwE43rKtkBjRIjVx1FjDfp65OyNuD7rRsPbYWxznCYKboPMTvJoAz9yuj25k-3BitWakblQVe3wrE8HXBNJt8IIvDAHurplRKaskaLjL66h_0OixxzPZWildM64Znih8oG0NKEbu7ZTgza8bmv4xzz8u_Xdx1_A41A6cHIEGPf8Y-rPgLB_GzFw</addsrcrecordid><sourcetype>Open Access Repository</sourcetype><iscdi>true</iscdi><recordtype>article</recordtype><pqid>2771309981</pqid></control><display><type>article</type><title>Support vector machine based emissions modeling using particle swarm optimization for homogeneous charge compression ignition engine</title><source>SAGE IMechE Complete Collection</source><source>SAGE</source><creator>Gordon, David ; Norouzi, Armin ; Blomeyer, Gero ; Bedei, Julian ; Aliramezani, Masoud ; Andert, Jakob ; Koch, Charles R</creator><creatorcontrib>Gordon, David ; Norouzi, Armin ; Blomeyer, Gero ; Bedei, Julian ; Aliramezani, Masoud ; Andert, Jakob ; Koch, Charles R</creatorcontrib><description>The internal combustion engine faces increasing societal and governmental pressure to improve both efficiency and engine out emissions. Currently, research has moved from traditional combustion methods to new highly efficient combustion strategies such as Homogeneous Charge Compression Ignition (HCCI). However, predicting the exact value of engine out emissions using conventional physics-based or data-driven models is still a challenge for engine researchers due to the complexity the of combustion and emission formation. Research has focused on using Artificial Neural Networks (ANN) for this problem but ANN’s require large training datasets for acceptable accuracy. This work addresses this problem by presenting the development of a simple model for predicting the steady-state emissions of a single cylinder HCCI engine which is created using an metaheuristic optimization based Support Vector Machine (SVM). The selection of input variables to the SVM model is explored using five different feature sets, considering up to seven engine inputs. The best results are achieved with a model combining linear and squared inputs as well as cross correlations and their squares totaling 26 features. In this case the model fit represented by R2 values were between 0.72 and 0.95. The best model fits were achieved for CO and CO2, while HC and NOx models have reduced model performance. Linear and non-linear SVM models were then compared to an ANN model. This comparison showed that SVM based models were more robust to changes in feature selection and better able to avoid local minimums compared to the ANN models leading to a more consistent model prediction when limited training data is available. The proposed machine learning based HCCI emission models and the feature selection approach provide insight into optimizing the model accuracy while minimizing the computational costs.</description><identifier>ISSN: 1468-0874</identifier><identifier>EISSN: 2041-3149</identifier><identifier>DOI: 10.1177/14680874211055546</identifier><identifier>PMID: 36776419</identifier><language>eng</language><publisher>London, England: SAGE Publications</publisher><subject>Artificial neural networks ; Cross correlation ; Emission ; Feature selection ; Heuristic methods ; Ignition ; Internal combustion engines ; Machine learning ; Model accuracy ; Particle swarm optimization ; Support vector machines ; Training</subject><ispartof>International journal of engine research, 2023-02, Vol.24 (2), p.536-551</ispartof><rights>IMechE 2021</rights><rights>IMechE 2021.</rights><rights>IMechE 2021 2021 Institution of Mechanical Engineers</rights><lds50>peer_reviewed</lds50><oa>free_for_read</oa><woscitedreferencessubscribed>false</woscitedreferencessubscribed><citedby>FETCH-LOGICAL-c466t-2e8890a924e4204460aa34a57315ea7dbadfaaef980f350449a59e17fcae05563</citedby><cites>FETCH-LOGICAL-c466t-2e8890a924e4204460aa34a57315ea7dbadfaaef980f350449a59e17fcae05563</cites><orcidid>0000-0002-7999-8234 ; 0000-0002-6754-1907 ; 0000-0001-8260-8754 ; 0000-0003-2690-0739 ; 0000-0002-6094-5933</orcidid></display><links><openurl>$$Topenurl_article</openurl><openurlfulltext>$$Topenurlfull_article</openurlfulltext><thumbnail>$$Tsyndetics_thumb_exl</thumbnail><linktopdf>$$Uhttps://journals.sagepub.com/doi/pdf/10.1177/14680874211055546$$EPDF$$P50$$Gsage$$Hfree_for_read</linktopdf><linktohtml>$$Uhttps://journals.sagepub.com/doi/10.1177/14680874211055546$$EHTML$$P50$$Gsage$$Hfree_for_read</linktohtml><link.rule.ids>230,314,780,784,885,21912,27923,27924,45058,45446,79135</link.rule.ids><backlink>$$Uhttps://www.ncbi.nlm.nih.gov/pubmed/36776419$$D View this record in MEDLINE/PubMed$$Hfree_for_read</backlink></links><search><creatorcontrib>Gordon, David</creatorcontrib><creatorcontrib>Norouzi, Armin</creatorcontrib><creatorcontrib>Blomeyer, Gero</creatorcontrib><creatorcontrib>Bedei, Julian</creatorcontrib><creatorcontrib>Aliramezani, Masoud</creatorcontrib><creatorcontrib>Andert, Jakob</creatorcontrib><creatorcontrib>Koch, Charles R</creatorcontrib><title>Support vector machine based emissions modeling using particle swarm optimization for homogeneous charge compression ignition engine</title><title>International journal of engine research</title><addtitle>Int J Engine Res</addtitle><description>The internal combustion engine faces increasing societal and governmental pressure to improve both efficiency and engine out emissions. Currently, research has moved from traditional combustion methods to new highly efficient combustion strategies such as Homogeneous Charge Compression Ignition (HCCI). However, predicting the exact value of engine out emissions using conventional physics-based or data-driven models is still a challenge for engine researchers due to the complexity the of combustion and emission formation. Research has focused on using Artificial Neural Networks (ANN) for this problem but ANN’s require large training datasets for acceptable accuracy. This work addresses this problem by presenting the development of a simple model for predicting the steady-state emissions of a single cylinder HCCI engine which is created using an metaheuristic optimization based Support Vector Machine (SVM). The selection of input variables to the SVM model is explored using five different feature sets, considering up to seven engine inputs. The best results are achieved with a model combining linear and squared inputs as well as cross correlations and their squares totaling 26 features. In this case the model fit represented by R2 values were between 0.72 and 0.95. The best model fits were achieved for CO and CO2, while HC and NOx models have reduced model performance. Linear and non-linear SVM models were then compared to an ANN model. This comparison showed that SVM based models were more robust to changes in feature selection and better able to avoid local minimums compared to the ANN models leading to a more consistent model prediction when limited training data is available. The proposed machine learning based HCCI emission models and the feature selection approach provide insight into optimizing the model accuracy while minimizing the computational costs.</description><subject>Artificial neural networks</subject><subject>Cross correlation</subject><subject>Emission</subject><subject>Feature selection</subject><subject>Heuristic methods</subject><subject>Ignition</subject><subject>Internal combustion engines</subject><subject>Machine learning</subject><subject>Model accuracy</subject><subject>Particle swarm optimization</subject><subject>Support vector machines</subject><subject>Training</subject><issn>1468-0874</issn><issn>2041-3149</issn><fulltext>true</fulltext><rsrctype>article</rsrctype><creationdate>2023</creationdate><recordtype>article</recordtype><sourceid>AFRWT</sourceid><recordid>eNp1kUFv1DAQhS0EotvCD-CCLHHhkmIndhxfkFAFBalSD4WzNetMsq7iONhJEZz54Ti7pVBQL_Zhvnkzbx4hLzg75VypN1zUDWuUKDlnUkpRPyKbkgleVFzox2Sz1osVOCLHKV0zxqRQ6ik5qmqlasH1hvy8WqYpxJneoJ1DpB7szo1It5CwpehdSi6MifrQ4uDGni5pfSeIs7MD0vQNoqdhmp13P2DOLO2yzC740OOIYUnU7iD2SG3wU8S9HHX96PYsjn2e9ow86WBI-Pz2PyFfPrz_fPaxuLg8_3T27qKwoq7nosSm0Qx0KVBkm6JmAJUAqSouEVS7hbYDwE43rKtkBjRIjVx1FjDfp65OyNuD7rRsPbYWxznCYKboPMTvJoAz9yuj25k-3BitWakblQVe3wrE8HXBNJt8IIvDAHurplRKaskaLjL66h_0OixxzPZWildM64Znih8oG0NKEbu7ZTgza8bmv4xzz8u_Xdx1_A41A6cHIEGPf8Y-rPgLB_GzFw</recordid><startdate>20230201</startdate><enddate>20230201</enddate><creator>Gordon, David</creator><creator>Norouzi, Armin</creator><creator>Blomeyer, Gero</creator><creator>Bedei, Julian</creator><creator>Aliramezani, Masoud</creator><creator>Andert, Jakob</creator><creator>Koch, Charles R</creator><general>SAGE Publications</general><general>SAGE PUBLICATIONS, INC</general><scope>AFRWT</scope><scope>NPM</scope><scope>AAYXX</scope><scope>CITATION</scope><scope>7TB</scope><scope>8FD</scope><scope>F28</scope><scope>FR3</scope><scope>7X8</scope><scope>5PM</scope><orcidid>https://orcid.org/0000-0002-7999-8234</orcidid><orcidid>https://orcid.org/0000-0002-6754-1907</orcidid><orcidid>https://orcid.org/0000-0001-8260-8754</orcidid><orcidid>https://orcid.org/0000-0003-2690-0739</orcidid><orcidid>https://orcid.org/0000-0002-6094-5933</orcidid></search><sort><creationdate>20230201</creationdate><title>Support vector machine based emissions modeling using particle swarm optimization for homogeneous charge compression ignition engine</title><author>Gordon, David ; Norouzi, Armin ; Blomeyer, Gero ; Bedei, Julian ; Aliramezani, Masoud ; Andert, Jakob ; Koch, Charles R</author></sort><facets><frbrtype>5</frbrtype><frbrgroupid>cdi_FETCH-LOGICAL-c466t-2e8890a924e4204460aa34a57315ea7dbadfaaef980f350449a59e17fcae05563</frbrgroupid><rsrctype>articles</rsrctype><prefilter>articles</prefilter><language>eng</language><creationdate>2023</creationdate><topic>Artificial neural networks</topic><topic>Cross correlation</topic><topic>Emission</topic><topic>Feature selection</topic><topic>Heuristic methods</topic><topic>Ignition</topic><topic>Internal combustion engines</topic><topic>Machine learning</topic><topic>Model accuracy</topic><topic>Particle swarm optimization</topic><topic>Support vector machines</topic><topic>Training</topic><toplevel>peer_reviewed</toplevel><toplevel>online_resources</toplevel><creatorcontrib>Gordon, David</creatorcontrib><creatorcontrib>Norouzi, Armin</creatorcontrib><creatorcontrib>Blomeyer, Gero</creatorcontrib><creatorcontrib>Bedei, Julian</creatorcontrib><creatorcontrib>Aliramezani, Masoud</creatorcontrib><creatorcontrib>Andert, Jakob</creatorcontrib><creatorcontrib>Koch, Charles R</creatorcontrib><collection>Sage Journals GOLD Open Access 2024</collection><collection>PubMed</collection><collection>CrossRef</collection><collection>Mechanical & Transportation Engineering Abstracts</collection><collection>Technology Research Database</collection><collection>ANTE: Abstracts in New Technology & Engineering</collection><collection>Engineering Research Database</collection><collection>MEDLINE - Academic</collection><collection>PubMed Central (Full Participant titles)</collection><jtitle>International journal of engine research</jtitle></facets><delivery><delcategory>Remote Search Resource</delcategory><fulltext>fulltext</fulltext></delivery><addata><au>Gordon, David</au><au>Norouzi, Armin</au><au>Blomeyer, Gero</au><au>Bedei, Julian</au><au>Aliramezani, Masoud</au><au>Andert, Jakob</au><au>Koch, Charles R</au><format>journal</format><genre>article</genre><ristype>JOUR</ristype><atitle>Support vector machine based emissions modeling using particle swarm optimization for homogeneous charge compression ignition engine</atitle><jtitle>International journal of engine research</jtitle><addtitle>Int J Engine Res</addtitle><date>2023-02-01</date><risdate>2023</risdate><volume>24</volume><issue>2</issue><spage>536</spage><epage>551</epage><pages>536-551</pages><issn>1468-0874</issn><eissn>2041-3149</eissn><abstract>The internal combustion engine faces increasing societal and governmental pressure to improve both efficiency and engine out emissions. Currently, research has moved from traditional combustion methods to new highly efficient combustion strategies such as Homogeneous Charge Compression Ignition (HCCI). However, predicting the exact value of engine out emissions using conventional physics-based or data-driven models is still a challenge for engine researchers due to the complexity the of combustion and emission formation. Research has focused on using Artificial Neural Networks (ANN) for this problem but ANN’s require large training datasets for acceptable accuracy. This work addresses this problem by presenting the development of a simple model for predicting the steady-state emissions of a single cylinder HCCI engine which is created using an metaheuristic optimization based Support Vector Machine (SVM). The selection of input variables to the SVM model is explored using five different feature sets, considering up to seven engine inputs. The best results are achieved with a model combining linear and squared inputs as well as cross correlations and their squares totaling 26 features. In this case the model fit represented by R2 values were between 0.72 and 0.95. The best model fits were achieved for CO and CO2, while HC and NOx models have reduced model performance. Linear and non-linear SVM models were then compared to an ANN model. This comparison showed that SVM based models were more robust to changes in feature selection and better able to avoid local minimums compared to the ANN models leading to a more consistent model prediction when limited training data is available. The proposed machine learning based HCCI emission models and the feature selection approach provide insight into optimizing the model accuracy while minimizing the computational costs.</abstract><cop>London, England</cop><pub>SAGE Publications</pub><pmid>36776419</pmid><doi>10.1177/14680874211055546</doi><tpages>16</tpages><orcidid>https://orcid.org/0000-0002-7999-8234</orcidid><orcidid>https://orcid.org/0000-0002-6754-1907</orcidid><orcidid>https://orcid.org/0000-0001-8260-8754</orcidid><orcidid>https://orcid.org/0000-0003-2690-0739</orcidid><orcidid>https://orcid.org/0000-0002-6094-5933</orcidid><oa>free_for_read</oa></addata></record> |
fulltext | fulltext |
identifier | ISSN: 1468-0874 |
ispartof | International journal of engine research, 2023-02, Vol.24 (2), p.536-551 |
issn | 1468-0874 2041-3149 |
language | eng |
recordid | cdi_pubmedcentral_primary_oai_pubmedcentral_nih_gov_9902987 |
source | SAGE IMechE Complete Collection; SAGE |
subjects | Artificial neural networks Cross correlation Emission Feature selection Heuristic methods Ignition Internal combustion engines Machine learning Model accuracy Particle swarm optimization Support vector machines Training |
title | Support vector machine based emissions modeling using particle swarm optimization for homogeneous charge compression ignition engine |
url | http://sfxeu10.hosted.exlibrisgroup.com/loughborough?ctx_ver=Z39.88-2004&ctx_enc=info:ofi/enc:UTF-8&ctx_tim=2025-01-12T10%3A48%3A26IST&url_ver=Z39.88-2004&url_ctx_fmt=infofi/fmt:kev:mtx:ctx&rfr_id=info:sid/primo.exlibrisgroup.com:primo3-Article-proquest_pubme&rft_val_fmt=info:ofi/fmt:kev:mtx:journal&rft.genre=article&rft.atitle=Support%20vector%20machine%20based%20emissions%20modeling%20using%20particle%20swarm%20optimization%20for%20homogeneous%20charge%20compression%20ignition%20engine&rft.jtitle=International%20journal%20of%20engine%20research&rft.au=Gordon,%20David&rft.date=2023-02-01&rft.volume=24&rft.issue=2&rft.spage=536&rft.epage=551&rft.pages=536-551&rft.issn=1468-0874&rft.eissn=2041-3149&rft_id=info:doi/10.1177/14680874211055546&rft_dat=%3Cproquest_pubme%3E2771309981%3C/proquest_pubme%3E%3Cgrp_id%3Ecdi_FETCH-LOGICAL-c466t-2e8890a924e4204460aa34a57315ea7dbadfaaef980f350449a59e17fcae05563%3C/grp_id%3E%3Coa%3E%3C/oa%3E%3Curl%3E%3C/url%3E&rft_id=info:oai/&rft_pqid=2771309981&rft_id=info:pmid/36776419&rft_sage_id=10.1177_14680874211055546&rfr_iscdi=true |