Joint prediction of observations and states in time-series based on belief functions - AGPIG Accéder directement au contenu
Article Dans Une Revue IEEE Transactions on Systems, Man, and Cybernetics, Part B: Cybernetics Année : 2013

Joint prediction of observations and states in time-series based on belief functions

Résumé

Forecasting the future states of a complex system is a complicated challenge that is encountered in many industrial applications covered in the community of Prognostics and Health Management (PHM). Practically, states can be either continuous or discrete: Continuous states generally represent the value of a signal while discrete states generally depict functioning modes reflecting the current degradation. For each case, specific techniques exist. In this paper, we propose an approach based on case-based reasoning that jointly estimates the future values of the continuous signal and the future discrete modes. The main characteristics of the proposed approach are the following: 1) It relies on the K-nearest neighbours algorithm based on belief functions theory; 2) Belief functions allow the user to represent his partial knowledge concerning the possible states in the training dataset, in particular concerning transitions between functioning modes which are imprecisely known; 3) Two distinct strategies are proposed for states prediction and the fusion of both strategies is also considered. Two real datasets were used in order to assess the performance in estimating future break-down of a real system.
Fichier principal
Vignette du fichier
EvKNN_pronostic_joint_prediction_discrete_continuous_belief_functions-1.pdf (753.18 Ko) Télécharger le fichier
Origine : Fichiers produits par l'(les) auteur(s)
Loading...

Dates et versions

hal-00719616 , version 1 (20-07-2012)

Identifiants

Citer

Emmanuel Ramasso, Michèle Rombaut, Noureddine Zerhouni. Joint prediction of observations and states in time-series based on belief functions. IEEE Transactions on Systems, Man, and Cybernetics, Part B: Cybernetics, 2013, 43 (1), pp.37-50. ⟨10.1109/TSMCB.2012.2198882⟩. ⟨hal-00719616⟩
612 Consultations
607 Téléchargements

Altmetric

Partager

Gmail Facebook X LinkedIn More