Modeling of biodiesel production using optimization designs from literature: aiming to reduce the laboratory workload

This study explores non-linear tree-based learning algorithms for modeling biodiesel reactions. A dataset of 3038 reaction samples from 111 published studies was compiled, each optimizing distinct biodiesel reaction systems. Key operational parameters were selected to represent the dataset's di...

Full description

Saved in:
Bibliographic Details
Main Authors: Iver Bergh Hvidsten, Kristian Hovde Liland, Oliver Tomic, Jorge Mario Marchetti
Format: Article
Language:English
Published: Elsevier 2025-10-01
Series:Fuel Processing Technology
Subjects:
Online Access:http://www.sciencedirect.com/science/article/pii/S037838202500089X
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1849689440335691776
author Iver Bergh Hvidsten
Kristian Hovde Liland
Oliver Tomic
Jorge Mario Marchetti
author_facet Iver Bergh Hvidsten
Kristian Hovde Liland
Oliver Tomic
Jorge Mario Marchetti
author_sort Iver Bergh Hvidsten
collection DOAJ
description This study explores non-linear tree-based learning algorithms for modeling biodiesel reactions. A dataset of 3038 reaction samples from 111 published studies was compiled, each optimizing distinct biodiesel reaction systems. Key operational parameters were selected to represent the dataset's diversity. Random forest (RF) and gradient boosting regressor (GBR) models were employed to predict biodiesel yield across the various reaction systems. GBR, with 1000 estimators and a tree depth of 5, achieved the best performance (R2 = 0.744, RMSE = 10.783). The global GBR model was comprehensively evaluated for accuracy and physical relevance, with proposed applications in component screening and reaction optimization using the DIRECT-l (DIviding RECTangles - locally biased version) algorithm. Additionally, an experimental reaction was optimized via the global model and DIRECT-l, then refined using a retrained local model for improved system-specific predictions. These models offer researchers a data-driven approach to selecting and optimizing biodiesel reactions, reducing laboratory time and improving predictive accuracy for specific systems.
format Article
id doaj-art-b96422cb2e844360b51ad054e301b146
institution DOAJ
issn 0378-3820
language English
publishDate 2025-10-01
publisher Elsevier
record_format Article
series Fuel Processing Technology
spelling doaj-art-b96422cb2e844360b51ad054e301b1462025-08-20T03:21:38ZengElsevierFuel Processing Technology0378-38202025-10-0127510826510.1016/j.fuproc.2025.108265Modeling of biodiesel production using optimization designs from literature: aiming to reduce the laboratory workloadIver Bergh Hvidsten0Kristian Hovde Liland1Oliver Tomic2Jorge Mario Marchetti3Faculty of Science and Technology, Realtek, Department of Physics, Norwegian University of Life Sciences, Ås, NorwayFaculty of Science and Technology, Realtek, Department of Industrial Economics & Technology Management, Norwegian University of Life Sciences, Ås, NorwayFaculty of Science and Technology, Realtek, Department of Data Science, Norwegian University of Life Sciences, Ås, NorwayFaculty of Science and Technology, Realtek, Department of Physics, Norwegian University of Life Sciences, Ås, Norway; Corresponding author at: Faculty of Science and Technology, Realtek, Norwegian University of Life Sciences, Drøbakveien 31, 1432 ÅS, Norway.This study explores non-linear tree-based learning algorithms for modeling biodiesel reactions. A dataset of 3038 reaction samples from 111 published studies was compiled, each optimizing distinct biodiesel reaction systems. Key operational parameters were selected to represent the dataset's diversity. Random forest (RF) and gradient boosting regressor (GBR) models were employed to predict biodiesel yield across the various reaction systems. GBR, with 1000 estimators and a tree depth of 5, achieved the best performance (R2 = 0.744, RMSE = 10.783). The global GBR model was comprehensively evaluated for accuracy and physical relevance, with proposed applications in component screening and reaction optimization using the DIRECT-l (DIviding RECTangles - locally biased version) algorithm. Additionally, an experimental reaction was optimized via the global model and DIRECT-l, then refined using a retrained local model for improved system-specific predictions. These models offer researchers a data-driven approach to selecting and optimizing biodiesel reactions, reducing laboratory time and improving predictive accuracy for specific systems.http://www.sciencedirect.com/science/article/pii/S037838202500089XBiodieselTransesterificationData miningGradient boosting regressorSHAPDIRECT-l
spellingShingle Iver Bergh Hvidsten
Kristian Hovde Liland
Oliver Tomic
Jorge Mario Marchetti
Modeling of biodiesel production using optimization designs from literature: aiming to reduce the laboratory workload
Fuel Processing Technology
Biodiesel
Transesterification
Data mining
Gradient boosting regressor
SHAP
DIRECT-l
title Modeling of biodiesel production using optimization designs from literature: aiming to reduce the laboratory workload
title_full Modeling of biodiesel production using optimization designs from literature: aiming to reduce the laboratory workload
title_fullStr Modeling of biodiesel production using optimization designs from literature: aiming to reduce the laboratory workload
title_full_unstemmed Modeling of biodiesel production using optimization designs from literature: aiming to reduce the laboratory workload
title_short Modeling of biodiesel production using optimization designs from literature: aiming to reduce the laboratory workload
title_sort modeling of biodiesel production using optimization designs from literature aiming to reduce the laboratory workload
topic Biodiesel
Transesterification
Data mining
Gradient boosting regressor
SHAP
DIRECT-l
url http://www.sciencedirect.com/science/article/pii/S037838202500089X
work_keys_str_mv AT iverberghhvidsten modelingofbiodieselproductionusingoptimizationdesignsfromliteratureaimingtoreducethelaboratoryworkload
AT kristianhovdeliland modelingofbiodieselproductionusingoptimizationdesignsfromliteratureaimingtoreducethelaboratoryworkload
AT olivertomic modelingofbiodieselproductionusingoptimizationdesignsfromliteratureaimingtoreducethelaboratoryworkload
AT jorgemariomarchetti modelingofbiodieselproductionusingoptimizationdesignsfromliteratureaimingtoreducethelaboratoryworkload