압력, RPM 데이터 및 LSTM을 이용한 디젤엔진의 이상탐지기법 연구

김만수¹^{, *} ; 최영원² ; 박병호² ; 박문성² ; 김진홍³

1LIG넥스원 미래전장IPS연구소 1팀 수석연구원
2LIG넥스원 미래전장IPS연구소 1팀 선임연구원
3LIG넥스원 미래전장IPS연구소 1팀 연구원

Study on Anomaly Detection in Diesel Engines Using Pressure Data, RPM Data and LSTM

Mansoo Kim¹^{, *} ; Youngwon Choi² ; Byeongho Bak² ; Moonsung Park² ; Jinhong Kim³

1Chief research engineer, Future Warfare Integrated Product Support R&D (Maritime), LIG Nex1
2Research engineer, Future Warfare Integrated Product Support R&D (Maritime), LIG Nex1
3Research engineer, Future Warfare Integrated Product Support R&D (Maritime), LIG Nex1

Correspondence to: ^*Mansoo Kim Future Warfare Integrated Product Support R&D(Maritime), LIG Nex1 21, Pangyo-ro 255beon-gil, Bundang-gu, Seongnam-si, Gyeonggi-do, Republic of Korea Tel: +82-31-326-9415 Fax: +82-31-288-9123 E-mail: mansoo.kim2@lignex1.com

초록

함정 디젤엔진의 고장을 사전에 탐지하는 것은 함정의 운항 안전을 보장하고 운영유지비용을 최소화하는 중요한 문제이다. 이를 위해 엔진에서 실시간으로 수집되는 각종 압력데이터와 RPM을 모니터링하여 이상을 조기에 감지하는 기술이 필요하다. 본 연구에서는 LSTM을 사용하여 함정의 엔진 고장을 사전에 탐지하는 방법론을 제안한다.

Abstract

Detecting failures in naval engines in advance is crucial for ensuring the safety of vessel operations and minimizing operational maintenance costs. To achieve this, it is essential to have technologies that can monitor various pressure data and RPM collected in real-time from the engine and detect anomalies early. This study proposes a methodology for predicting engine failures in naval vessels using Long Short-term Memory(LSTM).

Keywords:

Anomaly Detection, Long Short-term Memory, Condition Based Mainternance, Diesel Engine

키워드:

이상탐지, 장단기 메모리 네트워크, 상태기반정비, 디젤 엔진

References

Raptodimos, Y., Lazakis, I., Using Artificial Neural Network-self-organising Map for Data Clustering of Marine Engine Condition Monitoring Applications, Ships Offshore Strucrures., 13(6), 649-656, 2018 [https://doi.org/10.1080/17445302.2018.1443694]
Cai, C., Weng, X. & Zhang, C., A Novel Approach for Marine Diesel Engine Fault Diagnosis, Cluster Computing, 20, 1691-1702, 2017 [https://doi.org/10.1007/s10586-017-0748-0]
Li, Z., Yan, X., Guo, Z. et al. A New Intelligent Fusion Method of Multi-Dimensional Sensors and Its Application to Tribo-System Fault Diagnosis of Marine Diesel Engines, Tribology Letters, 47, 1-15, 2012 [https://doi.org/10.1007/s11249-012-9948-1]
Y. Zhan, Z. Shi and M. Liu, The Application of Support Vector Machines (SVM) to Fault Diagnosis of Marine Main Engine Cylinder Cover, IECON 2007 - 33rd Annual Conference of the IEEE Industrial Electronics Society, 3018-3022, 2007 [https://doi.org/10.1109/IECON.2007.4459891]
Wenkui Xi, Zhixiong Li, Zhe Tian, Zhihe Duan, A Feature Extraction and Visualization Method for Fault Detection of Marine Diesel Engines, Measurement, 116, 429-437, 2018 [https://doi.org/10.1016/j.measurement.2017.11.035]
Sang Gu Lee, Sung Woong Kim, Jae Young Lee, Hyeon Ju Jeong, Dong Hun Kang, A Cause Analysis Study on the Abnormal Vibration Characteristics of Diesel Engine Power Transmission System, Journal of the KNST, 5(1), 76-83, 2022 [https://doi.org/10.31818/JKNST.2022.03.5.1.76]
Sung-Hyun Kim, A Study of Machine Learning Technique for Noise-based Engine Fault Diagnosis, Journal of the KNST, 5(1), 16-19, 2022 [https://doi.org/10.31818/JKNST.2022.03.5.1.16]
Namoano Bernadin, Starr Andrew, Emmanouilidis Christos, Ruiz Carcel Cristobal, Engines Datasets, Cranfield Online Research Data (CORD), 2020, Dataset.