Towards a model-based condition assessment of complex marine machinery systems using systems engineering

Alexandros S. Zymaris, Øystein Åsheim Alnes, Knut Erik Knutsen, and Nikolaos M.P. Kakalis
Submission Type: 
Full Paper
phmec_16_032.pdf2.07 MBJune 23, 2016 - 4:46am

In the present paper, a systems engineering methodology is presented for the analysis and condition assessment of complex marine machinery systems. Two important characteristics of these systems are (i) that they comprise of a multitude of subcomponents that all of them influence the system condition and (ii) the continuously varying operating and environmental conditions. The methodology presented herein is capable to evaluate the system level effects of component degradation and faulty states under realistic system operation. By virtue of this, it is employed along with sensor signal data for the identification of degraded states and the allocation of the problem to specific system components. The modelling platform used in this work is the DNV▪GL COSSMOS. The latter provides the user with the capability to define and analyse a wide range of system configurations, model their behaviour in terms of their governing energy conversion processes (mechanical, electric, thermodynamic, heat transfer, fluid flow, etc.), and perform a number of model-based applications (simulation, optimisation, control, parameter estimation) under both steady-state (design/off-design conditions) and dynamic (time varying, transient operation) conditions. At the methodological level, automated model-based sensitivity analysis is conducted with respect to a set of component degradation/failure modes, the latter is used along with a clustering algorithm for the precise allocation of the failure to specific components and system particulars. The selected case-study is the diesel-electric marine propulsion system of a 3900 tonnes DWT anchor handling vessel embedded with its cooling network. Based on the results, the approach is capable to successfully identify faults at various subcomponents of the cooling network system including pumps, regulating valves, heat exchangers and piping. Due to the fact that the system is treated in an integrated manner, a fault can be identified in a component using sensor signals placed in other system locations. Moreover, comparisons between different sets of sensor signals and the corresponding method results are presented.

Publication Year: 
Publication Volume: 
Publication Control Number: 
Page Count: 
Submission Keywords: 
condition monitoring
systems engineering
Model-based diagnosis
sensitivity analysis
Marine Propuslion and cooling systems
Submission Topic Areas: 
Model-based methods for fault detection, diagnostics, and prognosis
Modeling and simulation
Systems and platform applications
Submitted by: 

follow us

PHM Society on Facebook Follow PHM Society on Twitter PHM Society on LinkedIn PHM Society RSS News Feed