Skip to Main content Skip to Navigation
Conference papers

Spatio-temporal convolutional neural networks for failure prediction

Abstract : The use of statistical learning techniques to identify a failure in a system by using time series collected from it is well known. However, in the case of an industrial system made of multiple subsystems, their direct application is limited by the system complexity. In the meantime, the application of those techniques individually to each subsystem does not take their dependencies into consideration leading to limited performances. The objective of this paper is to propose a model of spatio-temporal convolutional neural network able to consider spatial and temporal dependencies on time series collected on subsystems of an industrial system for failure classification.
Complete list of metadata

Cited literature [12 references]  Display  Hide  Download

https://hal.archives-ouvertes.fr/hal-02282219
Contributor : Yohan Petetin Connect in order to contact the contributor
Submitted on : Monday, September 9, 2019 - 7:07:07 PM
Last modification on : Friday, October 9, 2020 - 8:38:12 AM
Long-term archiving on: : Saturday, February 8, 2020 - 12:01:21 AM

File

gretsi_2019_YP.pdf
Files produced by the author(s)

Identifiers

  • HAL Id : hal-02282219, version 1

Citation

Nicolas Aussel, Fabian Dubourvieux, yohan Petetin. Spatio-temporal convolutional neural networks for failure prediction. GRETSI 2019: XXVIIème colloque francophone de traitement du signal et des images, Aug 2019, Lille, France. pp.1-5. ⟨hal-02282219⟩

Share

Metrics

Record views

104

Files downloads

96