N. Kempema, Conner Sharpe, Xiao Wu, Merhdad Shahabi, D. Kubinski
{"title":"Development of a Machine-Learning Classification Model for an\n Electrochemical Nitrogen Oxides Sensor in Gasoline Powertrains","authors":"N. Kempema, Conner Sharpe, Xiao Wu, Merhdad Shahabi, D. Kubinski","doi":"10.4271/03-16-04-0031","DOIUrl":null,"url":null,"abstract":"Future automotive emission regulations are becoming increasingly dependent on\n off-cycle (acquired on road and referred to as “real-world”) driving and\n testing. This was driven in part by the often-observed fact that laboratory\n emission drive cycles (developed to evaluate a vehicle’s emissions on a chassis\n dynamometer) may not fully capture the nature of real-world driving. As a\n result, portable emission measurement systems were developed that could be fit\n in the trunk of a vehicle, but were relatively large, expensive, and complex to\n operate. It would be advantageous to have low-cost and simple to operate\n on-board sensors that could be used in a gasoline powertrain to monitor\n important criteria emission species, such as NOx. The electrochemical\n NOx sensor is often used for emissions control systems in diesel\n powertrains and a proven technology for application to the relatively harsh\n environment of automotive exhaust. However, electrochemical NOx\n sensors are nearly equally sensitive to both NOx and NH3,\n setting up an implicit classification problem that must be solved before they\n can accurately measure NOx. In this work, we develop a\n machine-learning model to classify the output of a NOx sensor in a\n gasoline powertrain. A model generalization study is conducted, and the model is\n found to be ~96% accurate and able to predict NOx mass emitted over a\n drive cycle within ~9% of a perfectly classified NOx sensor.","PeriodicalId":47948,"journal":{"name":"SAE International Journal of Engines","volume":"14 1","pages":""},"PeriodicalIF":1.1000,"publicationDate":"2022-10-11","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":"0","resultStr":null,"platform":"Semanticscholar","paperid":null,"PeriodicalName":"SAE International Journal of Engines","FirstCategoryId":"1085","ListUrlMain":"https://doi.org/10.4271/03-16-04-0031","RegionNum":0,"RegionCategory":null,"ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":null,"EPubDate":"","PubModel":"","JCR":"Q3","JCRName":"TRANSPORTATION SCIENCE & TECHNOLOGY","Score":null,"Total":0}
引用次数: 0
Abstract
Future automotive emission regulations are becoming increasingly dependent on
off-cycle (acquired on road and referred to as “real-world”) driving and
testing. This was driven in part by the often-observed fact that laboratory
emission drive cycles (developed to evaluate a vehicle’s emissions on a chassis
dynamometer) may not fully capture the nature of real-world driving. As a
result, portable emission measurement systems were developed that could be fit
in the trunk of a vehicle, but were relatively large, expensive, and complex to
operate. It would be advantageous to have low-cost and simple to operate
on-board sensors that could be used in a gasoline powertrain to monitor
important criteria emission species, such as NOx. The electrochemical
NOx sensor is often used for emissions control systems in diesel
powertrains and a proven technology for application to the relatively harsh
environment of automotive exhaust. However, electrochemical NOx
sensors are nearly equally sensitive to both NOx and NH3,
setting up an implicit classification problem that must be solved before they
can accurately measure NOx. In this work, we develop a
machine-learning model to classify the output of a NOx sensor in a
gasoline powertrain. A model generalization study is conducted, and the model is
found to be ~96% accurate and able to predict NOx mass emitted over a
drive cycle within ~9% of a perfectly classified NOx sensor.