Advanced Detection and Monitoring System for JBT Marel’s GyroCompact
(2025) In CODEN:LUTEDX/TEIE EIEM01 20251Division for Industrial Electrical Engineering and Automation
- Abstract (Swedish)
- The food processing industry is integral to modern society, with increasing demands for efficiency and sustainability. Industrial food freezers, such as the JBT Marel’s GyroCompact spiral freezer, are energy-intensive systems operating for extended durations daily, making energy efficiency improvements critical both economically and environmentally.
This thesis aims to develop a vision-based monitoring and detection system for JBT Marel which could potentially be used to optimize freezer energy use by dynamically regulating power based on product load and distribution.
Current GyroCompact models lack automated detection systems, resulting in static power consumption irrespective of product throughput. By utilizing machine learning we... (More) - The food processing industry is integral to modern society, with increasing demands for efficiency and sustainability. Industrial food freezers, such as the JBT Marel’s GyroCompact spiral freezer, are energy-intensive systems operating for extended durations daily, making energy efficiency improvements critical both economically and environmentally.
This thesis aims to develop a vision-based monitoring and detection system for JBT Marel which could potentially be used to optimize freezer energy use by dynamically regulating power based on product load and distribution.
Current GyroCompact models lack automated detection systems, resulting in static power consumption irrespective of product throughput. By utilizing machine learning we propose implementing a cost-efficient object detection system solution to monitor food products entering the freezer in real time. RGB and infrared cameras were evaluated as potential sensors, considering factors like robustness, cost, and adaptability to industrial conditions.
While temperature monitoring and conventional computer vision techniques were explored using a thermal camera, it was deemed more difficult and more expensive than using a normal RGB-camera.
A proof-of-concept system was developed and tested using accessible hardware and software tools. Key questions addressed include the relevant parameters for monitoring, suitable object detection algorithms, and feasible hardware configurations. The evaluation emphasized the fact that its purpose would be as a demonstrator and early prototype. Ingress protection and food safety were not included in the proof-of-concept, and should JBTM wish to develop the prototype further those areas should be addressed.
Although limitations, such as customization for different freezer models, were acknowledged, the findings provide a robust foundation for future developments. (Less)
Please use this url to cite or link to this publication:
http://lup.lub.lu.se/student-papers/record/9212707
- author
- Fredriksson Kirrage, Tor LU and Sonnsjö Lönegren, Josef
- supervisor
- organization
- course
- EIEM01 20251
- year
- 2025
- type
- H3 - Professional qualifications (4 Years - )
- subject
- publication/series
- CODEN:LUTEDX/TEIE
- report number
- 5552
- language
- English
- id
- 9212707
- date added to LUP
- 2026-01-28 09:37:30
- date last changed
- 2026-01-28 09:37:30
@misc{9212707,
abstract = {{The food processing industry is integral to modern society, with increasing demands for efficiency and sustainability. Industrial food freezers, such as the JBT Marel’s GyroCompact spiral freezer, are energy-intensive systems operating for extended durations daily, making energy efficiency improvements critical both economically and environmentally.
This thesis aims to develop a vision-based monitoring and detection system for JBT Marel which could potentially be used to optimize freezer energy use by dynamically regulating power based on product load and distribution.
Current GyroCompact models lack automated detection systems, resulting in static power consumption irrespective of product throughput. By utilizing machine learning we propose implementing a cost-efficient object detection system solution to monitor food products entering the freezer in real time. RGB and infrared cameras were evaluated as potential sensors, considering factors like robustness, cost, and adaptability to industrial conditions.
While temperature monitoring and conventional computer vision techniques were explored using a thermal camera, it was deemed more difficult and more expensive than using a normal RGB-camera.
A proof-of-concept system was developed and tested using accessible hardware and software tools. Key questions addressed include the relevant parameters for monitoring, suitable object detection algorithms, and feasible hardware configurations. The evaluation emphasized the fact that its purpose would be as a demonstrator and early prototype. Ingress protection and food safety were not included in the proof-of-concept, and should JBTM wish to develop the prototype further those areas should be addressed.
Although limitations, such as customization for different freezer models, were acknowledged, the findings provide a robust foundation for future developments.}},
author = {{Fredriksson Kirrage, Tor and Sonnsjö Lönegren, Josef}},
language = {{eng}},
note = {{Student Paper}},
series = {{CODEN:LUTEDX/TEIE}},
title = {{Advanced Detection and Monitoring System for JBT Marel’s GyroCompact}},
year = {{2025}},
}