Fault Diagnosis System Based on Ontology for Fleet Case Reused

Voisin, Alexandre; Medina-Oliva, Gabriela; Monnin, Maxime; Léger, Jean-Baptiste; Iung, Benoît

doi:10.1007/978-3-319-15326-1_5

Cited by 6 publications

(3 citation statements)

References 30 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…• the heterogeneity of signals since systems evolve in different environments, with different missions and thus monitored signals show significant variations (Voisin, Medina-Oliva, Monnin, Léger, & Iung, 2013).…”

Section: Towards a Semantic Model Formalizationmentioning

confidence: 99%

Key factor identification for energy consumption analysis

Medina-Oliva,

Voisin,

Monnin

et al. 2014

PHME_CONF

View full text Add to dashboard Cite

Nowadays the economic, environmental and societal issues concerning energy consumption require a deeper understanding of the factors influencing it. The influencing factors could concern the technical characteristics of the systems, the operational conditions and usage of equipment, the environmental conditions, etc. To understand the main contributing factors a knowledge model with the influencing factors is formalized in the form of an ontology. This ontology model allows to distinguish in a general way the main concepts (i.e. factors) that show higher consumption trends. This way, a preliminary analysis reflecting the key influencing factors could be perform in order to focus later on a deeper analysis with data mining techniques. This paper focuses on the formalizationof an ontology model in the marine domain for energy consumption purposes. Then, the approach is illustrated with an example of a fleet of diesel engines.

show abstract

Section: Towards a Semantic Model Formalizationmentioning

confidence: 99%

Key factor identification for energy consumption analysis

Medina-Oliva,

Voisin,

Monnin

et al. 2014

PHME_CONF

View full text Add to dashboard Cite

show abstract

“…As a consequence solutions to fix problem can be more easily and quickly deployed over all the process measurement. The fleet dimension rely on the capacity to deal with similar/heterogeneous equipment (from sensor to large and complex equipment (e.g; engine…)), taking benefit form already developed approaches (Monnin et al 2011a, Voisin et al 2013.…”

Section: Outlook Of the Developed Approachmentioning

confidence: 99%

Data quality and reliability: a cornerstone for PHM processes

et al. 2014

View full text Add to dashboard Cite

In most of industrial processes, the measurement are central to the process control and quality management. This become even truer when measurement data are used to develop and support PHM strategies. In this context, many software are installed in order to collect data for providing quality assessment at each step of the manufacturing process. However, measurement error or drift are not considered leading to downgrading / rejected products / suboptimal running conditions that comes from measurement drift not detected on time. In concrete, these lead to bigger penalty than losses of production due to stopping time for repairing sensors. Indeed, generally speaking, process data is the “raw material” for many business processes, starting from process control strategy, PHM strategies to Business Intelligence. Thus being able to ensure data quality and reliability is of first importance. Towards this end, methods and tools are required to support online measurement monitoring, predictive diagnosis and reliability enhancement.In this paper, a dedicated approach developed in collaboration with ArcelorMittal Research is presented. It consists in the development of intelligent software that would enable sensor measurement validation taking into account process parameters and operational conditions. An illustrative case study is extracted from an ongoing application developed for the finishing line in ArcelorMittal plant at Florange in France. Results regarding measurement reliability assessment as well as sensor failure anticipation will be described.

show abstract

“…In (Monnin, Voisin, Leger, & Iung, 2011) and (Voisin, Medina-Oliva, Monnin, Leger, & Iung, 2013) a knowledge structuring scheme for achieving fleet-wide PHM utilizing both modelling and monitoring is proposed. Utilizing contextual information about the system and components by means of semantic modelling is identified as a key issue in order to allow for consideration of fleet component similarities and heterogeneities.…”

Section: Introductionmentioning

confidence: 99%

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

Zymaris¹,

Alnes

Knutsen

et al. 2016

PHME_CONF

View full text Add to dashboard Cite

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 which influence the overall system condition/performance 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 DNVGL COSSMOS (Complex Ship System Modelling & Simulation).At the methodological level, an 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 2300 tonnes DWT (deadweight tonnage) 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.

show abstract

Fault Diagnosis System Based on Ontology for Fleet Case Reused

Cited by 6 publications

References 30 publications

Key factor identification for energy consumption analysis

Key factor identification for energy consumption analysis

Data quality and reliability: a cornerstone for PHM processes

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

Contact Info

Product

Resources

About