A machine learning based quality control system for power cable manufacturing

Jussi Hanhirova; Jaakko Harjuhahto; Janne Harjuhahto; Vesa Hirvisalo

doi:10.1109/INDIN41052.2019.8972281

A machine learning based quality control system for power cable manufacturing

Department of Computer Science

Maillefer Extrusion Oy

Research output: Chapter in Book/Report/Conference proceeding › Conference article in proceedings › Scientific › peer-review

7 Citations (Scopus)

1867 Downloads (Pure)

Abstract

We study methods for observing physical defects on the surface of power cables. Quality control is essential for power cable manufacturing and surface defects are an important quality factor. Traditionally power cable manufacturing has relied on manual inspection as automated methods have not been sufficient to be used in industrial production. We have designed and implemented a novel defect detection system that applies machine learning methods to detect power cable surface defects. Our system uses laser scanning to map the surface of a cable during production. For the machine learning, we have evaluated different CNN (Convolutional Neural Net- work) architectures and studied their performance and accuracy. According to our results, CNNs are suitable for the detection of surface defects as they can be trained with large amounts of cable surface data.

Original language	English
Title of host publication	Proceedings - 2019 IEEE 17th International Conference on Industrial Informatics, INDIN 2019
Publisher	IEEE
Pages	193-198
Number of pages	6
ISBN (Electronic)	978-1-7281-2927-3
DOIs	https://doi.org/10.1109/INDIN41052.2019.8972281
Publication status	Published - 2019
MoE publication type	A4 Conference publication
Event	IEEE International Conference on Industrial Informatics - Aalto University, Helsinki and Espoo, Finland Duration: 22 Jul 2019 → 25 Jul 2019 Conference number: 17 https://www.indin2019.org/

Publication series

Name	IEEE International Conference on Industrial Informatics
ISSN (Electronic)	2378-363X

Conference

Conference	IEEE International Conference on Industrial Informatics
Abbreviated title	INDIN
Country/Territory	Finland
City	Helsinki and Espoo
Period	22/07/2019 → 25/07/2019
Internet address	https://www.indin2019.org/

UN SDGs

This output contributes to the following UN Sustainable Development Goals (SDGs)

SDG 9 Industry, Innovation, and Infrastructure

Keywords

deep learning
convolutional neural networks, power cables, quality controlvolutional neural networks
power cables
quality control

Access to Document

10.1109/INDIN41052.2019.8972281

SCI_Hanhirova_A machine learning based quality control system for power cable manufacturingAccepted author manuscript, 691 KB

Cite this

Hanhirova, J., Harjuhahto, J., Harjuhahto, J., & Hirvisalo, V. (2019). A machine learning based quality control system for power cable manufacturing. In Proceedings - 2019 IEEE 17th International Conference on Industrial Informatics, INDIN 2019 (pp. 193-198). Article 8972281 (IEEE International Conference on Industrial Informatics). IEEE. https://doi.org/10.1109/INDIN41052.2019.8972281

@inproceedings{78c1e0d771a64d10805c6744e6bcdd2d,

title = "A machine learning based quality control system for power cable manufacturing",

abstract = "We study methods for observing physical defects on the surface of power cables. Quality control is essential for power cable manufacturing and surface defects are an important quality factor. Traditionally power cable manufacturing has relied on manual inspection as automated methods have not been sufficient to be used in industrial production. We have designed and implemented a novel defect detection system that applies machine learning methods to detect power cable surface defects. Our system uses laser scanning to map the surface of a cable during production. For the machine learning, we have evaluated different CNN (Convolutional Neural Net- work) architectures and studied their performance and accuracy. According to our results, CNNs are suitable for the detection of surface defects as they can be trained with large amounts of cable surface data.",

keywords = "deep learning, convolutional neural networks, power cables, quality controlvolutional neural networks, power cables, quality control",

author = "Jussi Hanhirova and Jaakko Harjuhahto and Janne Harjuhahto and Vesa Hirvisalo",

year = "2019",

doi = "10.1109/INDIN41052.2019.8972281",

language = "English",

series = "IEEE International Conference on Industrial Informatics",

publisher = "IEEE",

pages = "193--198",

booktitle = "Proceedings - 2019 IEEE 17th International Conference on Industrial Informatics, INDIN 2019",

address = "United States",

note = "IEEE International Conference on Industrial Informatics, INDIN ; Conference date: 22-07-2019 Through 25-07-2019",

url = "https://www.indin2019.org/",

}

Hanhirova, J , Harjuhahto, J, Harjuhahto, J & Hirvisalo, V 2019, A machine learning based quality control system for power cable manufacturing. in Proceedings - 2019 IEEE 17th International Conference on Industrial Informatics, INDIN 2019., 8972281, IEEE International Conference on Industrial Informatics, IEEE, pp. 193-198, IEEE International Conference on Industrial Informatics, Helsinki and Espoo, Finland, 22/07/2019. https://doi.org/10.1109/INDIN41052.2019.8972281

A machine learning based quality control system for power cable manufacturing. / Hanhirova, Jussi ; Harjuhahto, Jaakko; Harjuhahto, Janne et al.
Proceedings - 2019 IEEE 17th International Conference on Industrial Informatics, INDIN 2019. IEEE, 2019. p. 193-198 8972281 (IEEE International Conference on Industrial Informatics).

Research output: Chapter in Book/Report/Conference proceeding › Conference article in proceedings › Scientific › peer-review

TY - GEN

T1 - A machine learning based quality control system for power cable manufacturing

AU - Hanhirova, Jussi

AU - Harjuhahto, Jaakko

AU - Harjuhahto, Janne

AU - Hirvisalo, Vesa

N1 - Conference code: 17

PY - 2019

Y1 - 2019

N2 - We study methods for observing physical defects on the surface of power cables. Quality control is essential for power cable manufacturing and surface defects are an important quality factor. Traditionally power cable manufacturing has relied on manual inspection as automated methods have not been sufficient to be used in industrial production. We have designed and implemented a novel defect detection system that applies machine learning methods to detect power cable surface defects. Our system uses laser scanning to map the surface of a cable during production. For the machine learning, we have evaluated different CNN (Convolutional Neural Net- work) architectures and studied their performance and accuracy. According to our results, CNNs are suitable for the detection of surface defects as they can be trained with large amounts of cable surface data.

AB - We study methods for observing physical defects on the surface of power cables. Quality control is essential for power cable manufacturing and surface defects are an important quality factor. Traditionally power cable manufacturing has relied on manual inspection as automated methods have not been sufficient to be used in industrial production. We have designed and implemented a novel defect detection system that applies machine learning methods to detect power cable surface defects. Our system uses laser scanning to map the surface of a cable during production. For the machine learning, we have evaluated different CNN (Convolutional Neural Net- work) architectures and studied their performance and accuracy. According to our results, CNNs are suitable for the detection of surface defects as they can be trained with large amounts of cable surface data.

KW - deep learning

KW - convolutional neural networks, power cables, quality controlvolutional neural networks

KW - power cables

KW - quality control

U2 - 10.1109/INDIN41052.2019.8972281

DO - 10.1109/INDIN41052.2019.8972281

M3 - Conference article in proceedings

T3 - IEEE International Conference on Industrial Informatics

SP - 193

EP - 198

BT - Proceedings - 2019 IEEE 17th International Conference on Industrial Informatics, INDIN 2019

PB - IEEE

T2 - IEEE International Conference on Industrial Informatics

Y2 - 22 July 2019 through 25 July 2019

ER -

A machine learning based quality control system for power cable manufacturing

Abstract

Publication series

Conference

UN SDGs

Keywords

Access to Document

Fingerprint

Cite this