Training the Recurrent neural network by the Fuzzy Min-Max algorithm for fault prediction

Journal Article · Wed Mar 04 23:00:00 EST 2009 · AIP Conference Proceedings

DOI:https://doi.org/10.1063/1.3106518· OSTI ID:21293340

Zemouri, Ryad ^[1]; Racoceanu, Daniel ^[2]; Zerhouni, Noureddine ^[3]; Minca, Eugenia ^[4]; Filip, Florin ^[5]

---

1. Laboratoire d'automatique, CNAM, 21 rue Pinel, 75013 Paris (France)
2. IPAL, UMI CNRS 2955, UJF, I2R/A-STAR, NUS, 1 Fusionopolis Way, 21-01 Connexis, 138632 Singapore (Singapore)
3. FEMTO-ST-UMR CNRS 6174, ENSMM, UFC, UTBM, 32 Avenue de l'Observatoire, 25044 Besancon (France)
4. Faculty of Electric Engineering, Valahia University, Bd. Unirii, nr. 18, 0200, Targoviste (Romania)
5. Romanian Academy, Calea Victoriei 125, Sct. 1, Bucuresti (Romania)

In this paper, we present a training technique of a Recurrent Radial Basis Function neural network for fault prediction. We use the Fuzzy Min-Max technique to initialize the k-center of the RRBF neural network. The k-means algorithm is then applied to calculate the centers that minimize the mean square error of the prediction task. The performances of the k-means algorithm are then boosted by the Fuzzy Min-Max technique.

• OSTI ID: 21293340
• Journal Information: AIP Conference Proceedings, Journal Name: AIP Conference Proceedings Journal Issue: 1 Vol. 1107; ISSN APCPCS; ISSN 0094-243X
• Country of Publication: United States
• Language: English

Related Subjects

97 MATHEMATICS AND COMPUTING
ALGORITHMS
CONTROL
ERRORS
FUZZY LOGIC
NEURAL NETWORKS
PATTERN RECOGNITION
PERFORMANCE
SYSTEM FAILURE ANALYSIS