Integrated design of prognosis, diagnosis and monitoring processes for proactive maintenance of manufacturing systems

Today, conventional maintenance strategies such as corrective and systematic ones are not sufficient to fulfil the industrial needs on maximum reduction of failures and degradations of manufacturing systems. Indeed as these strategies are made sometimes too early or too late, it is required to evolve towards just in time strategies such as proactive one. This proactive maintenance strategy aims at increasing instantaneous availability, mean availability and safety of the industrial systems. Its originality is based on the prognosis of the degradation through a prognosis process enabling to propagate the degradation cause in order to analyse the new degraded situation and to anticipate the manufacturing system failure. This cause is identified by a diagnosis process determining the cause origin from the symptom observations of other degradation highlighted by a monitoring process. So, this paper aims at presenting our contribution to the proactive maintenance processes modelling within a formal framework based on GERAM and CIMOSA work and integrating a maintenance specialist semantics. This contribution is validated through an implementation of proactive maintenance systems for particular production systems: turbines of two hydroelectric power stations in the context of the European ESPRIT REMAFEX no.20874 and PRIMA no.20775 projects.

[1]  Dean Allemang,et al.  The Computational Complexity of Abduction , 1991, Artif. Intell..

[2]  Kurt Kosanke,et al.  Enterprise Engineering and Integration: Building International Consensus, Proceedings of the International Conference on Enterprise Integration and Modeling Technology, ICEIMT 1997, Torino, Italy, October 28-30, 1997 , 1997, International Conference on Enterprise Integration and Modeling Technique.

[3]  Marcel Staroswiecki,et al.  Duality of Analytical Redundancy and Statistical Approach in Fault Diagnosis , 1996 .

[4]  Alan S. Willsky,et al.  A survey of design methods for failure detection in dynamic systems , 1976, Autom..

[5]  Pietro Torasso,et al.  A spectrum of logical definitions of model‐based diagnosis 1 , 1991, Comput. Intell..

[6]  Brian C. Williams,et al.  Diagnosing Multiple Faults , 1987, Artif. Intell..

[7]  Spyros G. Tzafestas,et al.  Industrial Forecasting Using Knowledge-Based Techniques and Artificial Neural Networks , 1999 .

[8]  Benoît Iung,et al.  An innovative approach for new distributed maintenance system: application to hydro power plants of the REMAFEX project , 1999 .

[9]  G. Morel,et al.  Distributed intelligent actuators and sensors , 1993, 1993 CompEuro Proceedings Computers in Design, Manufacturing, and Production.

[10]  H. G Lawley.  Operability Studies and Hazard Analysis , 1974 .

[11]  Rolf Isermann,et al.  Trends in the Application of Model Based Fault Detection and Diagnosis of Technical Processes , 1996 .

[12]  J. Zaytoon,et al.  A contribution for integrating the operational safety concept in CIM , 1992 .

[13]  R. Keith Mobley,et al.  An introduction to predictive maintenance , 1989 .

[14]  J.J. Gertler,et al.  Survey of model-based failure detection and isolation in complex plants , 1988, IEEE Control Systems Magazine.