Staff View: Predicting biodiesel properties and its optimal fatty acid profile via explainable machine learning

Predicting biodiesel properties and its optimal fatty acid profile via explainable machine learning

The accurate prediction of biodiesel fuel properties and determination of its optimal fatty acid (FA) profiles is a non-trivial process. To this aim, machine learning (ML) based predictive models were developed for cetane number (CN) and cold filter plugging point (CFPP), where the extreme gradient...

Full description

Saved in:

Bibliographic Details
Main Authors:	Manu Suvarna, Mohammad Islam Jahirul, Yeap, Aaron Wai Hung, Cheryl Valencia Augustine, Anushri Umesh, Mohammad Golam Rasul, Mehmet Erdem Günay, Ramazan Yildirim, Jidon Janaun
Format:	Article
Language:	English English
Published:	Elsevier 2022
Subjects:	TA1-2040 Engineering (General). Civil engineering (General) TP315-360 Fuel
Online Access:	https://eprints.ums.edu.my/id/eprint/33803/1/Predicting%20biodiesel%20properties%20and%20its%20optimal%20fatty%20acid%20profile%20via%20explainable%20machine%20learning.pdf https://eprints.ums.edu.my/id/eprint/33803/2/Predicting%20biodiesel%20properties%20and%20its%20optimal%20fatty%20acid%20profile%20via%20explainable%20machine%20learning1.pdf https://eprints.ums.edu.my/id/eprint/33803/ https://www.sciencedirect.com/science/article/pii/S0960148122002737?casa_token=jZDbLUKi4vAAAAAA:TMsOj_09hXypyKOx2c-DEzQUJK7b9HGwssbT15h3i2nVWNzbnNiiqK9ybWYMn_L6pvYxXOK3W6M https://doi.org/10.1016/j.renene.2022.02.124
Tags:	Add Tag No Tags, Be the first to tag this record!

id	my.ums.eprints.33803
record_format	eprints
spelling	my.ums.eprints.338032022-08-16T06:55:55Z https://eprints.ums.edu.my/id/eprint/33803/ Predicting biodiesel properties and its optimal fatty acid profile via explainable machine learning Manu Suvarna Mohammad Islam Jahirul Yeap, Aaron Wai Hung Cheryl Valencia Augustine Anushri Umesh Mohammad Golam Rasul Mehmet Erdem Günay Ramazan Yildirim Jidon Janaun TA1-2040 Engineering (General). Civil engineering (General) TP315-360 Fuel The accurate prediction of biodiesel fuel properties and determination of its optimal fatty acid (FA) profiles is a non-trivial process. To this aim, machine learning (ML) based predictive models were developed for cetane number (CN) and cold filter plugging point (CFPP), where the extreme gradient boost (XGB) and random forest (RF) algorithms had the best performance with R2 of 0.89 and 0.91 on the test data, respectively. A classifier model for oxidative stability (OS) was devised to predict if it would pass or fail the ASTM and EU limits, where the support vector classifier (SVC) had the highest accuracy of 0.93 and 0.77 for ASTM and EU limits. Causal analysis via Shapley and Accumulated Local Effects revealed the significance and correlation of FAs with the fuel properties. This eventually aided the determination of the optimal FA composition via evolutionary optimization, such that the properties would meet the ASTM and EU standards. This study presents an end-to-end ML framework including descriptive, predictive, causal and prescriptive analytics to predict biodiesel fuel properties as a function of its FA composition; and eventually prescribes the optimal FA composition necessary to ensure that the fuel properties meet the regulatory standards. Elsevier 2022 Article PeerReviewed text en https://eprints.ums.edu.my/id/eprint/33803/1/Predicting%20biodiesel%20properties%20and%20its%20optimal%20fatty%20acid%20profile%20via%20explainable%20machine%20learning.pdf text en https://eprints.ums.edu.my/id/eprint/33803/2/Predicting%20biodiesel%20properties%20and%20its%20optimal%20fatty%20acid%20profile%20via%20explainable%20machine%20learning1.pdf Manu Suvarna and Mohammad Islam Jahirul and Yeap, Aaron Wai Hung and Cheryl Valencia Augustine and Anushri Umesh and Mohammad Golam Rasul and Mehmet Erdem Günay and Ramazan Yildirim and Jidon Janaun (2022) Predicting biodiesel properties and its optimal fatty acid profile via explainable machine learning. Renewable Energy, 189. pp. 245-258. ISSN 0960-1481 https://www.sciencedirect.com/science/article/pii/S0960148122002737?casa_token=jZDbLUKi4vAAAAAA:TMsOj_09hXypyKOx2c-DEzQUJK7b9HGwssbT15h3i2nVWNzbnNiiqK9ybWYMn_L6pvYxXOK3W6M https://doi.org/10.1016/j.renene.2022.02.124
institution	Universiti Malaysia Sabah
building	UMS Library
collection	Institutional Repository
continent	Asia
country	Malaysia
content_provider	Universiti Malaysia Sabah
content_source	UMS Institutional Repository
url_provider	http://eprints.ums.edu.my/
language	English English
topic	TA1-2040 Engineering (General). Civil engineering (General) TP315-360 Fuel
spellingShingle	TA1-2040 Engineering (General). Civil engineering (General) TP315-360 Fuel Manu Suvarna Mohammad Islam Jahirul Yeap, Aaron Wai Hung Cheryl Valencia Augustine Anushri Umesh Mohammad Golam Rasul Mehmet Erdem Günay Ramazan Yildirim Jidon Janaun Predicting biodiesel properties and its optimal fatty acid profile via explainable machine learning
description	The accurate prediction of biodiesel fuel properties and determination of its optimal fatty acid (FA) profiles is a non-trivial process. To this aim, machine learning (ML) based predictive models were developed for cetane number (CN) and cold filter plugging point (CFPP), where the extreme gradient boost (XGB) and random forest (RF) algorithms had the best performance with R2 of 0.89 and 0.91 on the test data, respectively. A classifier model for oxidative stability (OS) was devised to predict if it would pass or fail the ASTM and EU limits, where the support vector classifier (SVC) had the highest accuracy of 0.93 and 0.77 for ASTM and EU limits. Causal analysis via Shapley and Accumulated Local Effects revealed the significance and correlation of FAs with the fuel properties. This eventually aided the determination of the optimal FA composition via evolutionary optimization, such that the properties would meet the ASTM and EU standards. This study presents an end-to-end ML framework including descriptive, predictive, causal and prescriptive analytics to predict biodiesel fuel properties as a function of its FA composition; and eventually prescribes the optimal FA composition necessary to ensure that the fuel properties meet the regulatory standards.
format	Article
author	Manu Suvarna Mohammad Islam Jahirul Yeap, Aaron Wai Hung Cheryl Valencia Augustine Anushri Umesh Mohammad Golam Rasul Mehmet Erdem Günay Ramazan Yildirim Jidon Janaun
author_facet	Manu Suvarna Mohammad Islam Jahirul Yeap, Aaron Wai Hung Cheryl Valencia Augustine Anushri Umesh Mohammad Golam Rasul Mehmet Erdem Günay Ramazan Yildirim Jidon Janaun
author_sort	Manu Suvarna
title	Predicting biodiesel properties and its optimal fatty acid profile via explainable machine learning
title_short	Predicting biodiesel properties and its optimal fatty acid profile via explainable machine learning
title_full	Predicting biodiesel properties and its optimal fatty acid profile via explainable machine learning
title_fullStr	Predicting biodiesel properties and its optimal fatty acid profile via explainable machine learning
title_full_unstemmed	Predicting biodiesel properties and its optimal fatty acid profile via explainable machine learning
title_sort	predicting biodiesel properties and its optimal fatty acid profile via explainable machine learning
publisher	Elsevier
publishDate	2022
url	https://eprints.ums.edu.my/id/eprint/33803/1/Predicting%20biodiesel%20properties%20and%20its%20optimal%20fatty%20acid%20profile%20via%20explainable%20machine%20learning.pdf https://eprints.ums.edu.my/id/eprint/33803/2/Predicting%20biodiesel%20properties%20and%20its%20optimal%20fatty%20acid%20profile%20via%20explainable%20machine%20learning1.pdf https://eprints.ums.edu.my/id/eprint/33803/ https://www.sciencedirect.com/science/article/pii/S0960148122002737?casa_token=jZDbLUKi4vAAAAAA:TMsOj_09hXypyKOx2c-DEzQUJK7b9HGwssbT15h3i2nVWNzbnNiiqK9ybWYMn_L6pvYxXOK3W6M https://doi.org/10.1016/j.renene.2022.02.124
_version_	1760231210945085440
score	13.211869

Predicting biodiesel properties and its optimal fatty acid profile via explainable machine learning

Similar Items