A Vibration-Based Machine Learning Approach for Roller Gap Detection in Biomass Pellet Production

Mads Kjærgaard Nielsen; Simon Klinge Nielsen; Torben Tambo

doi:10.1016/j.biosystemseng.2024.11.007

A Vibration-Based Machine Learning Approach for Roller Gap Detection in Biomass Pellet Production

Mads Kjærgaard Nielsen, Simon Klinge Nielsen, Torben Tambo^*

^*Corresponding author for this work

Department of Business Development and Technology

Research output: Contribution to journal/Conference contribution in journal/Contribution to newspaper › Journal article › Research › peer-review

Abstract

This research focuses on optimising biomass pellet manufacturing processes by detecting roller gap variations in rotary ring die pelleting (RRDP) technology. Integrating experimental testing, response surface modelling (RSM), and vibration-based machine learning, this study aims to ensure optimal conditions for biomass pellet mill operation. Vibration-based machine learning techniques offer an approach for detecting roller gap variations, while RSM provides mathematical models to understand process dynamics for identifying optimisation criteria. Experimental testing explores the impact of process variables on pellet quality metrics. Results demonstrate machine learning model performance in detecting roller gap variations with F1-scores ranging from 88.1% to 100.0% across a pilot- and industrial-scaled setup. ANOVA results underscore significant relationships between roller gap, feedstock layer mass, and pelleting process metrics, while the created RSM models all have determination coefficients R² of ≥0.90. Overall, this comprehensive approach contributes valuable insights into optimising efficiency and product quality in the biomass industry through an integrated framework.

Original language	English
Journal	Biosystems Engineering
Volume	248
Pages (from-to)	283-296
Number of pages	14
ISSN	1537-5110
DOIs	https://doi.org/10.1016/j.biosystemseng.2024.11.007
Publication status	Published - Dec 2024

Keywords

biomass pellets
rotary ring die pelletiser
machine learning
response surface methodology
vibration analysis

Access to Document

10.1016/j.biosystemseng.2024.11.007Licence: CC BY

Cite this

@article{81c4076c9d0e4014a14de90f15ba9aaa,

title = "A Vibration-Based Machine Learning Approach for Roller Gap Detection in Biomass Pellet Production",

abstract = "This research focuses on optimising biomass pellet manufacturing processes by detecting roller gap variations in rotary ring die pelleting (RRDP) technology. Integrating experimental testing, response surface modelling (RSM), and vibration-based machine learning, this study aims to ensure optimal conditions for biomass pellet mill operation. Vibration-based machine learning techniques offer an approach for detecting roller gap variations, while RSM provides mathematical models to understand process dynamics for identifying optimisation criteria. Experimental testing explores the impact of process variables on pellet quality metrics. Results demonstrate machine learning model performance in detecting roller gap variations with F1-scores ranging from 88.1% to 100.0% across a pilot- and industrial-scaled setup. ANOVA results underscore significant relationships between roller gap, feedstock layer mass, and pelleting process metrics, while the created RSM models all have determination coefficients R2 of ≥0.90. Overall, this comprehensive approach contributes valuable insights into optimising efficiency and product quality in the biomass industry through an integrated framework.",

keywords = "biomass pellets, rotary ring die pelletiser, machine learning, response surface methodology, vibration analysis",

author = "Nielsen, {Mads Kj{\ae}rgaard} and Nielsen, {Simon Klinge} and Torben Tambo",

year = "2024",

month = dec,

doi = "10.1016/j.biosystemseng.2024.11.007",

language = "English",

volume = "248",

pages = "283--296",

journal = "Biosystems Engineering",

issn = "1537-5110",

publisher = "Academic Press",

}

A Vibration-Based Machine Learning Approach for Roller Gap Detection in Biomass Pellet Production. / Nielsen, Mads Kjærgaard; Nielsen, Simon Klinge; Tambo, Torben.
In: Biosystems Engineering, Vol. 248, 12.2024, p. 283-296.

Research output: Contribution to journal/Conference contribution in journal/Contribution to newspaper › Journal article › Research › peer-review

TY - JOUR

T1 - A Vibration-Based Machine Learning Approach for Roller Gap Detection in Biomass Pellet Production

AU - Nielsen, Mads Kjærgaard

AU - Nielsen, Simon Klinge

AU - Tambo, Torben

PY - 2024/12

Y1 - 2024/12

N2 - This research focuses on optimising biomass pellet manufacturing processes by detecting roller gap variations in rotary ring die pelleting (RRDP) technology. Integrating experimental testing, response surface modelling (RSM), and vibration-based machine learning, this study aims to ensure optimal conditions for biomass pellet mill operation. Vibration-based machine learning techniques offer an approach for detecting roller gap variations, while RSM provides mathematical models to understand process dynamics for identifying optimisation criteria. Experimental testing explores the impact of process variables on pellet quality metrics. Results demonstrate machine learning model performance in detecting roller gap variations with F1-scores ranging from 88.1% to 100.0% across a pilot- and industrial-scaled setup. ANOVA results underscore significant relationships between roller gap, feedstock layer mass, and pelleting process metrics, while the created RSM models all have determination coefficients R2 of ≥0.90. Overall, this comprehensive approach contributes valuable insights into optimising efficiency and product quality in the biomass industry through an integrated framework.

AB - This research focuses on optimising biomass pellet manufacturing processes by detecting roller gap variations in rotary ring die pelleting (RRDP) technology. Integrating experimental testing, response surface modelling (RSM), and vibration-based machine learning, this study aims to ensure optimal conditions for biomass pellet mill operation. Vibration-based machine learning techniques offer an approach for detecting roller gap variations, while RSM provides mathematical models to understand process dynamics for identifying optimisation criteria. Experimental testing explores the impact of process variables on pellet quality metrics. Results demonstrate machine learning model performance in detecting roller gap variations with F1-scores ranging from 88.1% to 100.0% across a pilot- and industrial-scaled setup. ANOVA results underscore significant relationships between roller gap, feedstock layer mass, and pelleting process metrics, while the created RSM models all have determination coefficients R2 of ≥0.90. Overall, this comprehensive approach contributes valuable insights into optimising efficiency and product quality in the biomass industry through an integrated framework.

KW - biomass pellets

KW - rotary ring die pelletiser

KW - machine learning

KW - response surface methodology

KW - vibration analysis

UR - http://www.scopus.com/inward/record.url?scp=85209230763&partnerID=8YFLogxK

U2 - 10.1016/j.biosystemseng.2024.11.007

DO - 10.1016/j.biosystemseng.2024.11.007

M3 - Journal article

SN - 1537-5110

VL - 248

SP - 283

EP - 296

JO - Biosystems Engineering

JF - Biosystems Engineering

ER -

A Vibration-Based Machine Learning Approach for Roller Gap Detection in Biomass Pellet Production

Abstract

Keywords

Access to Document

Other files and links

Fingerprint

Cite this