Prognostic by classification of predictions combining similarity-based estimation and belief functions

Emmanuel Ramasso (1) , Michèle Rombaut (2) , Noureddine Zerhouni (1)

1 FEMTO-ST - Franche-Comté Électronique Mécanique, Thermique et Optique - Sciences et Technologies (UMR 6174)
2 GIPSA-AGPIG - GIPSA - Architecture, Géométrie, Perception, Images, Gestes

Forecasting the future states of a complex system is of paramount importance in many industrial applications covered in the community of Prognostics and Health Management (PHM). Practically, states can be either continuous (the value of a signal) or discrete (functioning modes). For each case, specific techniques exist. In this paper, we propose an approach called EVIPRO-KNN based on case-based reasoning and belief functions that jointly estimates the future values of the continuous signal and of the future discrete modes. A real datasets is used in order to assess the performance in estimating future break-down of a real system where the combination of both strategies provide the best prediction accuracies, up to 90%.

Domains: Automatic

Emmanuel Ramasso, Michèle Rombaut, Noureddine Zerhouni. Prognostic by classification of predictions combining similarity-based estimation and belief functions. T. Denoeux & M.H. Masson. Belief Functions : Theory and Applications, AISC 164. Proceedings of the 2nd International Conference on Belif Functions. Compiègne - 9-11 May 2012, 164, Springer-Verlag Berlin Heidelberg 2012, pp.61-68, 2012, Advances in Intelligent and Soft Computing. Volume 164 2012, 978-3-642-29460-0. ⟨10.1007/978-3-642-29461-77⟩. ⟨hal-00719583⟩