Predicting xylose yield from prehydrolysis of hardwoods: A machine learning approach
Hemicelluloses are amorphous polymers of sugar molecules that make up a major fraction of lignocellulosic biomasses. They have applications in the bioenergy, textile, mining, cosmetic, and pharmaceutical industries. Industrial use of hemicellulose often requires that the polymer be hydrolyzed into c...
| Main Authors: | , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
Frontiers Media S.A.
2022-10-01
|
| Series: | Frontiers in Chemical Engineering |
| Subjects: | |
| Online Access: | https://www.frontiersin.org/articles/10.3389/fceng.2022.994428/full |
| _version_ | 1797995832510775296 |
|---|---|
| author | Edward Wang Riley Ballachay Genpei Cai Yankai Cao Heather L. Trajano Heather L. Trajano |
| author_facet | Edward Wang Riley Ballachay Genpei Cai Yankai Cao Heather L. Trajano Heather L. Trajano |
| author_sort | Edward Wang |
| collection | DOAJ |
| description | Hemicelluloses are amorphous polymers of sugar molecules that make up a major fraction of lignocellulosic biomasses. They have applications in the bioenergy, textile, mining, cosmetic, and pharmaceutical industries. Industrial use of hemicellulose often requires that the polymer be hydrolyzed into constituent oligomers and monomers. Traditional models of hemicellulose degradation are kinetic, and usually only appropriate for limited operating regimes and specific species. The study of hemicellulose hydrolysis has yielded substantial data in the literature, enabling a diverse data set to be collected for general and widely applicable machine learning models. In this paper, a dataset containing 1955 experimental data points on batch hemicellulose hydrolysis of hardwood was collected from 71 published papers dated from 1985 to 2019. Three machine learning models (ridge regression, support vector regression and artificial neural networks) are assessed on their ability to predict xylose yield and compared to a kinetic model. Although the performance of ridge regression was unsatisfactory, both support vector regression and artificial neural networks outperformed the simple kinetic model. The artificial neural network outperformed support vector regression, reducing the mean absolute error in predicting soluble xylose yield of test data to 6.18%. The results suggest that machine learning models trained on historical data may be used to supplement experimental data, reducing the number of experiments needed. |
| first_indexed | 2024-04-11T10:07:57Z |
| format | Article |
| id | doaj.art-fa033e5c76094f39916201f2ef44dab0 |
| institution | Directory Open Access Journal |
| issn | 2673-2718 |
| language | English |
| last_indexed | 2024-04-11T10:07:57Z |
| publishDate | 2022-10-01 |
| publisher | Frontiers Media S.A. |
| record_format | Article |
| series | Frontiers in Chemical Engineering |
| spelling | doaj.art-fa033e5c76094f39916201f2ef44dab02022-12-22T04:30:11ZengFrontiers Media S.A.Frontiers in Chemical Engineering2673-27182022-10-01410.3389/fceng.2022.994428994428Predicting xylose yield from prehydrolysis of hardwoods: A machine learning approachEdward Wang0Riley Ballachay1Genpei Cai2Yankai Cao3Heather L. Trajano4Heather L. Trajano5Department of Chemical and Biological Engineering, The University of British Columbia, Vancouver, BC, CanadaDepartment of Chemical and Biological Engineering, The University of British Columbia, Vancouver, BC, CanadaDepartment of Chemical and Biological Engineering, The University of British Columbia, Vancouver, BC, CanadaDepartment of Chemical and Biological Engineering, The University of British Columbia, Vancouver, BC, CanadaDepartment of Chemical and Biological Engineering, The University of British Columbia, Vancouver, BC, CanadaBioProducts Institute, The University of British Columbia, Vancouver, BC, CanadaHemicelluloses are amorphous polymers of sugar molecules that make up a major fraction of lignocellulosic biomasses. They have applications in the bioenergy, textile, mining, cosmetic, and pharmaceutical industries. Industrial use of hemicellulose often requires that the polymer be hydrolyzed into constituent oligomers and monomers. Traditional models of hemicellulose degradation are kinetic, and usually only appropriate for limited operating regimes and specific species. The study of hemicellulose hydrolysis has yielded substantial data in the literature, enabling a diverse data set to be collected for general and widely applicable machine learning models. In this paper, a dataset containing 1955 experimental data points on batch hemicellulose hydrolysis of hardwood was collected from 71 published papers dated from 1985 to 2019. Three machine learning models (ridge regression, support vector regression and artificial neural networks) are assessed on their ability to predict xylose yield and compared to a kinetic model. Although the performance of ridge regression was unsatisfactory, both support vector regression and artificial neural networks outperformed the simple kinetic model. The artificial neural network outperformed support vector regression, reducing the mean absolute error in predicting soluble xylose yield of test data to 6.18%. The results suggest that machine learning models trained on historical data may be used to supplement experimental data, reducing the number of experiments needed.https://www.frontiersin.org/articles/10.3389/fceng.2022.994428/fullhemicellulosedilute acid hydrolysisautohydrolysiskineticsmachine learningsupport vector regression |
| spellingShingle | Edward Wang Riley Ballachay Genpei Cai Yankai Cao Heather L. Trajano Heather L. Trajano Predicting xylose yield from prehydrolysis of hardwoods: A machine learning approach Frontiers in Chemical Engineering hemicellulose dilute acid hydrolysis autohydrolysis kinetics machine learning support vector regression |
| title | Predicting xylose yield from prehydrolysis of hardwoods: A machine learning approach |
| title_full | Predicting xylose yield from prehydrolysis of hardwoods: A machine learning approach |
| title_fullStr | Predicting xylose yield from prehydrolysis of hardwoods: A machine learning approach |
| title_full_unstemmed | Predicting xylose yield from prehydrolysis of hardwoods: A machine learning approach |
| title_short | Predicting xylose yield from prehydrolysis of hardwoods: A machine learning approach |
| title_sort | predicting xylose yield from prehydrolysis of hardwoods a machine learning approach |
| topic | hemicellulose dilute acid hydrolysis autohydrolysis kinetics machine learning support vector regression |
| url | https://www.frontiersin.org/articles/10.3389/fceng.2022.994428/full |
| work_keys_str_mv | AT edwardwang predictingxyloseyieldfromprehydrolysisofhardwoodsamachinelearningapproach AT rileyballachay predictingxyloseyieldfromprehydrolysisofhardwoodsamachinelearningapproach AT genpeicai predictingxyloseyieldfromprehydrolysisofhardwoodsamachinelearningapproach AT yankaicao predictingxyloseyieldfromprehydrolysisofhardwoodsamachinelearningapproach AT heatherltrajano predictingxyloseyieldfromprehydrolysisofhardwoodsamachinelearningapproach AT heatherltrajano predictingxyloseyieldfromprehydrolysisofhardwoodsamachinelearningapproach |