{"title":"Spectroscopy-Based Machine Learning Approach to Predict Engine Fuel\n Properties of Biodiesel","authors":"Kiran Raj Bukkarapu, Anand Krishnasamy","doi":"10.4271/03-17-07-0051","DOIUrl":null,"url":null,"abstract":"Various feedstocks can be employed for biodiesel production, leading to\n considerable variation in composition and engine fuel characteristics. Using\n biodiesels originating from diverse feedstocks introduces notable variations in\n engine characteristics. Therefore, it is imperative to scrutinize the\n composition and properties of biodiesel before deployment in engines, a task\n facilitated by predictive models. Additionally, the international\n commercialization of biodiesel fuel is contingent upon stringent regulations.\n The traditional experimental measurement of biodiesel properties is laborious\n and expensive, necessitating skilled personnel. Predictive models offer an\n alternative approach by estimating biodiesel properties without depending on\n experimental measurements. This research is centered on building models that\n correlate mid-infrared spectra of biodiesel and critical fuel properties,\n encompassing kinematic viscosity, cetane number, and calorific value. The\n novelty of this investigation lies in exploring the suitability of support\n vector machine (SVM) regression, a burgeoning machine learning algorithm, for\n developing these models. Hyperparameter optimization for the SVM models was\n conducted using the grid search method, Bayesian optimization, and gray wolf\n optimization algorithms. The resultant SVM models exhibited a noteworthy\n reduction in mean absolute percentage error (MAPE) for the prediction of\n biodiesel viscosity (3.1%), cetane number (3%), and calorific value (2.1%). SVM\n regression, thus, emerges as a proficient machine learning algorithm capable of\n establishing correlations between the mid-infrared spectra of biodiesel and its\n properties, facilitating the reliable prediction of biodiesel\n characteristics.","PeriodicalId":47948,"journal":{"name":"SAE International Journal of Engines","volume":null,"pages":null},"PeriodicalIF":1.1000,"publicationDate":"2024-04-11","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/10.4271/03-17-07-0051","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
Various feedstocks can be employed for biodiesel production, leading to
considerable variation in composition and engine fuel characteristics. Using
biodiesels originating from diverse feedstocks introduces notable variations in
engine characteristics. Therefore, it is imperative to scrutinize the
composition and properties of biodiesel before deployment in engines, a task
facilitated by predictive models. Additionally, the international
commercialization of biodiesel fuel is contingent upon stringent regulations.
The traditional experimental measurement of biodiesel properties is laborious
and expensive, necessitating skilled personnel. Predictive models offer an
alternative approach by estimating biodiesel properties without depending on
experimental measurements. This research is centered on building models that
correlate mid-infrared spectra of biodiesel and critical fuel properties,
encompassing kinematic viscosity, cetane number, and calorific value. The
novelty of this investigation lies in exploring the suitability of support
vector machine (SVM) regression, a burgeoning machine learning algorithm, for
developing these models. Hyperparameter optimization for the SVM models was
conducted using the grid search method, Bayesian optimization, and gray wolf
optimization algorithms. The resultant SVM models exhibited a noteworthy
reduction in mean absolute percentage error (MAPE) for the prediction of
biodiesel viscosity (3.1%), cetane number (3%), and calorific value (2.1%). SVM
regression, thus, emerges as a proficient machine learning algorithm capable of
establishing correlations between the mid-infrared spectra of biodiesel and its
properties, facilitating the reliable prediction of biodiesel
characteristics.