Implementation of a self-learning algorithm for main engine condition monitoring

Research output: Chapter in Book/Report/Conference proceedingChapter

Abstract

Ship machinery maintenance can be seen as a convoluted optimisation problem where managerial, economic, and technical aspects are considered for vessel operation to remain sustainable and profitable. Recent literature shows that condition monitoring of ship systems has been tackled on single and independent component level. However, there has recently been a research tendency towards holistic system modelling. In this respect, this paper presents a methodology for intelligent, system-level modelling for the monitoring of main engine performance utilising data acquired through noon-reports. The proposed methodology will train a main engine expected-performance model. Training is based on one-class Support Vector Machine (SVM) classifier. Newly-acquired data are compared against model output and the probability of belonging to the same performance profile as used for model training is estimated. This will eventually lead to increasing ship operability and income through operational enhancement and minimisation of ship downtime. Two case studies are included, one utilising main engine data and one diesel generator data. Both are complemented by a sensitivity analysis, showing successful results in the recognition of deviant, abnormal ship machinery conditions.
LanguageEnglish
Title of host publicationMaritime Transportation and Harvesting of Sea Resources
Subtitle of host publicationProceedings of the 17th International Congress of the International Maritime Association of the Mediterranean (IMAM 2017)
EditorsC. Guedes Soares, Angelo P. Teixeira
Pages981-989
Number of pages8
Volume2
Publication statusPublished - 9 Jul 2017
Event17th International Congress of the International Maritime Association of the Mediterranean, Lisbon, Portugal, 9-11 October 2017 - Lisbon, Portugal
Duration: 9 Oct 201711 Oct 2017
Conference number: 17

Conference

Conference17th International Congress of the International Maritime Association of the Mediterranean, Lisbon, Portugal, 9-11 October 2017
Abbreviated titleIMAM 2017
CountryPortugal
CityLisbon
Period9/10/1711/10/17

Fingerprint

Condition monitoring
Learning algorithms
Ships
Engines
Machinery
Intelligent systems
Sensitivity analysis
Support vector machines
Classifiers
Economics
Monitoring

Keywords

  • ship machinery maintenance
  • system modelling
  • engine performance

Cite this

Gkerekos, C., Lazakis, I., & Theotokatos, G. (2017). Implementation of a self-learning algorithm for main engine condition monitoring. In C. G. Soares, & A. P. Teixeira (Eds.), Maritime Transportation and Harvesting of Sea Resources: Proceedings of the 17th International Congress of the International Maritime Association of the Mediterranean (IMAM 2017) (Vol. 2, pp. 981-989)
Gkerekos, Christos ; Lazakis, Iraklis ; Theotokatos, Gerasimos. / Implementation of a self-learning algorithm for main engine condition monitoring. Maritime Transportation and Harvesting of Sea Resources: Proceedings of the 17th International Congress of the International Maritime Association of the Mediterranean (IMAM 2017). editor / C. Guedes Soares ; Angelo P. Teixeira. Vol. 2 2017. pp. 981-989
@inbook{5f44736dfd44434c97ee20d7402df691,
title = "Implementation of a self-learning algorithm for main engine condition monitoring",
abstract = "Ship machinery maintenance can be seen as a convoluted optimisation problem where managerial, economic, and technical aspects are considered for vessel operation to remain sustainable and profitable. Recent literature shows that condition monitoring of ship systems has been tackled on single and independent component level. However, there has recently been a research tendency towards holistic system modelling. In this respect, this paper presents a methodology for intelligent, system-level modelling for the monitoring of main engine performance utilising data acquired through noon-reports. The proposed methodology will train a main engine expected-performance model. Training is based on one-class Support Vector Machine (SVM) classifier. Newly-acquired data are compared against model output and the probability of belonging to the same performance profile as used for model training is estimated. This will eventually lead to increasing ship operability and income through operational enhancement and minimisation of ship downtime. Two case studies are included, one utilising main engine data and one diesel generator data. Both are complemented by a sensitivity analysis, showing successful results in the recognition of deviant, abnormal ship machinery conditions.",
keywords = "ship machinery maintenance, system modelling, engine performance",
author = "Christos Gkerekos and Iraklis Lazakis and Gerasimos Theotokatos",
year = "2017",
month = "7",
day = "9",
language = "English",
isbn = "978-0-8153-7993-5",
volume = "2",
pages = "981--989",
editor = "Soares, {C. Guedes} and Teixeira, {Angelo P.}",
booktitle = "Maritime Transportation and Harvesting of Sea Resources",

}

Gkerekos, C, Lazakis, I & Theotokatos, G 2017, Implementation of a self-learning algorithm for main engine condition monitoring. in CG Soares & AP Teixeira (eds), Maritime Transportation and Harvesting of Sea Resources: Proceedings of the 17th International Congress of the International Maritime Association of the Mediterranean (IMAM 2017). vol. 2, pp. 981-989, 17th International Congress of the International Maritime Association of the Mediterranean, Lisbon, Portugal, 9-11 October 2017, Lisbon, Portugal, 9/10/17.

Implementation of a self-learning algorithm for main engine condition monitoring. / Gkerekos, Christos; Lazakis, Iraklis; Theotokatos, Gerasimos.

Maritime Transportation and Harvesting of Sea Resources: Proceedings of the 17th International Congress of the International Maritime Association of the Mediterranean (IMAM 2017). ed. / C. Guedes Soares; Angelo P. Teixeira. Vol. 2 2017. p. 981-989.

Research output: Chapter in Book/Report/Conference proceedingChapter

TY - CHAP

T1 - Implementation of a self-learning algorithm for main engine condition monitoring

AU - Gkerekos, Christos

AU - Lazakis, Iraklis

AU - Theotokatos, Gerasimos

PY - 2017/7/9

Y1 - 2017/7/9

N2 - Ship machinery maintenance can be seen as a convoluted optimisation problem where managerial, economic, and technical aspects are considered for vessel operation to remain sustainable and profitable. Recent literature shows that condition monitoring of ship systems has been tackled on single and independent component level. However, there has recently been a research tendency towards holistic system modelling. In this respect, this paper presents a methodology for intelligent, system-level modelling for the monitoring of main engine performance utilising data acquired through noon-reports. The proposed methodology will train a main engine expected-performance model. Training is based on one-class Support Vector Machine (SVM) classifier. Newly-acquired data are compared against model output and the probability of belonging to the same performance profile as used for model training is estimated. This will eventually lead to increasing ship operability and income through operational enhancement and minimisation of ship downtime. Two case studies are included, one utilising main engine data and one diesel generator data. Both are complemented by a sensitivity analysis, showing successful results in the recognition of deviant, abnormal ship machinery conditions.

AB - Ship machinery maintenance can be seen as a convoluted optimisation problem where managerial, economic, and technical aspects are considered for vessel operation to remain sustainable and profitable. Recent literature shows that condition monitoring of ship systems has been tackled on single and independent component level. However, there has recently been a research tendency towards holistic system modelling. In this respect, this paper presents a methodology for intelligent, system-level modelling for the monitoring of main engine performance utilising data acquired through noon-reports. The proposed methodology will train a main engine expected-performance model. Training is based on one-class Support Vector Machine (SVM) classifier. Newly-acquired data are compared against model output and the probability of belonging to the same performance profile as used for model training is estimated. This will eventually lead to increasing ship operability and income through operational enhancement and minimisation of ship downtime. Two case studies are included, one utilising main engine data and one diesel generator data. Both are complemented by a sensitivity analysis, showing successful results in the recognition of deviant, abnormal ship machinery conditions.

KW - ship machinery maintenance

KW - system modelling

KW - engine performance

UR - https://www.crcpress.com/Developments-in-Maritime-Transportation-and-Harvesting-of-Sea-Resources/Soares-Teixeira/p/book/9780815379935

M3 - Chapter

SN - 978-0-8153-7993-5

VL - 2

SP - 981

EP - 989

BT - Maritime Transportation and Harvesting of Sea Resources

A2 - Soares, C. Guedes

A2 - Teixeira, Angelo P.

ER -

Gkerekos C, Lazakis I, Theotokatos G. Implementation of a self-learning algorithm for main engine condition monitoring. In Soares CG, Teixeira AP, editors, Maritime Transportation and Harvesting of Sea Resources: Proceedings of the 17th International Congress of the International Maritime Association of the Mediterranean (IMAM 2017). Vol. 2. 2017. p. 981-989