Anomaly Detection Using Dynamical Linear Models and Sequential Testing on a Marine Engine System

##plugins.themes.bootstrap3.article.main##

##plugins.themes.bootstrap3.article.sidebar##

Published Oct 2, 2017
Erik Vanem Geir Olve Storvik

Abstract

This paper presents a study on the use of Dynamical Linear Models for anomaly detection and condition monitoring of a marine engine system. Various sensors are installed at different places within the engine system and records essential parameters such as power output from the engine, engine speed, bearing temperatures and various other temperatures, speeds and pressures for selected engine components. The idea is to utilize the information in these sensor signals in order to monitor the condition of the engine. Such a condition monitoring system should include means of fault detection, diagnosis and prognostics, where robust anomaly detection is a prerequisite for reliable management of the system. Dynamical
Linear Models (DLM) constitute a flexible framework for modelling of sensor signals, where the sensor signals are modelled conditional on some latent states, and the model provides forecasts of the signals that can be compared to new sensor readings. Statistical sequential model testing will then be performed on the forecast errors and model breakdown can be an indication of deviation from normal conditions and possible
impending failures of the engine system. This will then call for further diagnostics and prognostics tasks to interpret the nature of the deviations. The Dynamical Linear Model framework can accommodate a range of candidate models. However, very complicated models in high dimensions may be computationally expensive to estimate and apply, so various pre-processing techniques are investigated in this paper to improve model performance, including simple regression models, cluster analysis and principal component transformation.

How to Cite

Vanem, E., & Storvik, G. O. (2017). Anomaly Detection Using Dynamical Linear Models and Sequential Testing on a Marine Engine System. Annual Conference of the PHM Society, 9(1). https://doi.org/10.36001/phmconf.2017.v9i1.2397
Abstract 268 | PDF Downloads 189

##plugins.themes.bootstrap3.article.details##

Keywords

Condition monitoring, Time series analysis, Marine applications, Fault detection, Data driven models

References
Akaike, H. (1974). A new look at the statistical model identification. IEEE Transactions on Automatic Control, 19, 716-723.
Baraldi, P., Di Maio, F., Pappaglione, L., Zio, E., & Seraoui, R. (2012). Conditional monitoring of electrical power plant components during operational transients. Proceedings of the Institution of Mechanical Engineers, Part O: Journal of Risk and Reliability, 226, 568-583.
Brandsæter, A., Manno, G., Vanem, E., & Glad, I. K. (2016, June). An application of sensor based anomaly detection in the maritime industry. In Proc. IEEE PHM2016.
DNV GL. (2015). Ship connectivity (Tech. Rep. No. Strategic Research and Innovation Position Paper 4-2015). DNV GL.
DNV GL. (2017a). Rules for classification - general regulations. (DNVGL-RU-0050)
DNV GL. (2017b). Rules for classification: Ships. (DNVGLRU-SHIP)
Eleftheria, E., Papanikolaou, A., & Voulgarellis, M. (2016). Statistical analysis of ship accidents and review of safety level. Safety Science, 85, 282-292.
Hayes, M. A., & Capretz, M. A. (2015). Contextual anomaly detection framework for big sensor data. Journal of Big Data, 2, 1-22.
Hines, J. W., & Garvey, D. R. (2006). Development and application of fault detectability performance metrics for instrument calibration verification and anomaly detection. Journal of Pattern Recognition Research, 1, 2-15.
IMO. (2014a). International Safety Management Code (ISM Code) with guidelines for its implementation (2014 edition). International Maritime Organization.
IMO. (2014b). SOLAS, consolidated edition 2014. International Maritime Organization.
Kulldorff, M., Davis, R. L., Kolczak, Lewis, E., Lieu, T., & Platt, R. (2011). A maximized sequential probability ratio test for drug and vaccine safety surveillance. Sequential Analysis, 30, 58-78.
Mindykowski, J., & Tarasiuk, T. (2015). Problems of power quality in the wake of ship technology development. Ocean Engineering, 107, 108-117.
Niculita, O., Nwora, O., & Skaf, Z. (2017). Towards design of prognostics and health management solutions for maritime assets. Procedia CIRP, 59, 122-132.
Psarros, G. A. (2015, May-June). Comparing the navigator’s response time in collision and grounding accidents. In Proc. 34th international conference on ocean, offshore and arctic engineering (omae 2015).
Song, X., Wu, M., Jermaine, C., & Ranka, S. (2007). Conditional anomaly detection. IEEE Transactions on Knowledge and Data Engineering, 19, 631-645.
Vanem, E., Brandsæter, A., & Gramstad, O. (2016, September). Regression models for the effect of environmental conditions on the efficiency of ship machinery systems. In Proc. esrel 2016.
Vanem, E., Rus°as, S., Skjong, R., & Olufsen, O. (2007). Collision damage stability of passenger ships: Holistic and risk-based approach. International Shipbuilding Progress, 54, 323-337.
Vanem, E., & Skjong, R. (2004, July). Fire and evacuation risk assessment for passenger ships. In Proc. 10th international fire science and engineering conference (interflam) 2004 (Vol. 1, p. 365-374).
Vanem, E., & Storvik, G. O. (2016, August). Dynamical linear models for condition monitoring with multivariate sensor data. In Proc. comadem 2016.
Wald, A. (1945). Sequential tests of statistical hypotheses. The Annals of Mathematical Statistics, 16, 117-186.
West, M., & Harrison, J. (1997). Bayesian forecasting and dynamic models (Second ed.). Springer-Verlag.
Zymaris, A. S., Alnes, Ø. A° ., Knutsen, K. E., & Kakalis, N. M. P. (2016, July). Towards a model-based condition assessment of complex marine machinery systems using systems engineering. In Proc. Third European Conference of the Prognostics and Health Management Society 2016.
Section
Technical Research Papers