앙상블 기법을 이용한 선박 메인엔진 빅데이터의 이상치 탐지
Outlier detection of main engine data of a ship using ensemble method
1
조회
0
다운로드
6
인용
0
참고문헌
키워드
초록
본 논문은 선박 메인엔진 빅데이터를 비지도 학습 분석을 통해 메인 엔진 주요 부품의 고장 유무를 진단할 수 있는 기계학습 기반의 이상치 탐지 모델을 제안한다. 이를 위해 약 7개월 이상 선박 메인엔진 빅데이터를 수집하고, 메인 엔진 운전에 긴밀한 관계가 있는 특징 선택을 위해 전문가지식과 상관관계 분석을 수행하였다. 비지도 학습 분석은 다양한 알고리즘을 적용하여 선박 메인엔진의 운전 데이터를 학습하고, 실선 운영시 메인엔진의 이상이 식별할 수 있도록 구성 및 연구하였다.
Abstract
This paper proposes an outlier detection model based on machine learning that can diagnose the presence or absence of major engine parts through unsupervised learning analysis of main engine big data of a ship. Engine big data of the ship was collected for more than seven months, and expert knowledge and correlation analysis were performed to select features that are closely related to the operation of the main engine. For unsupervised learning analysis, ensemble model wherein many predictive models are strategically combined to increase the model performance, is used for anomaly detection. As a result, the proposed model successfully detected the anomalous engine status from the normal status. To validate our approach, clustering analysis was conducted to find out the different patterns of anomalies the anomalous point. By examining distribution of each cluster, we could successfully find the patterns of anomalies.
참고문헌 (11)
- Charu C. Aggarwal, "Outlier ensembles", ACM SIGKDD Explorations Newsletter, 14(2), pp.49~58 (2013), Association for Computing Machinery (ACM)[DOI]
- Aldous L, "Uncertainty analysis in ship performance monitoring", Ocean Engineering, 110, pp.29~38 (2015)[DOI]
- 배영목, "다변량 관리도를 활용한 선박 메인 엔진의 이상 관리 상한선 결정에 관한 연구", 대한조선학회 논문집, 55(6), pp.505~513 (2018), 대한조선학회[DOI]
- Deris S, "Ship maintenance scheduling by genetic algorithm and constraint-based reasoning", European Journal of Operational Research, 112, pp.489~502 (1999)[DOI]
- 김경엽, "이상감지 상관계수를 이용한선박디젤기관의 고장진단시스템에 관한 연구", 한국항행학회논문지, 15(1), pp.18~24 (2011), 한국항행학회
- Lazakis I, "Increasing ship operational reliability through the implementation of a holistic maintenance management strategy", Ships and Offshore Structures, 5, pp.337~357 (2010)[DOI]
- Lundberg SM, "A unified approach to interpreting model predictions", Proceedings of advances in neural information processing systems 2017, pp.4765~4774 (2017)
- Mak L, "Ship performance monitoring and analysis to improve fuel efficiency", Proceedings of 2014 Oceans St. John's, pp.1~10 (2014)
- Perera LP, "Marine engine-centered data analytics for ship performance monitoring", Journal of Offshore Mechanics and Arctic Engineering, 139, pp.1~28 (2017)[DOI]
- Phaladiganon P, "Bootstrap-based T2 multivariate control charts", Communications in Statistics-Simulation and Computation, 40, pp.645~662 (2011)[DOI]
- Soner O, "Use of tree based methods in ship performance monitoring under operating conditions", Ocean Engineering, 166, pp.302~310 (2018)[DOI]