Deep Variational Autoencoder: An Efficient Tool for PHM Frameworks

Ryad Zemouri (1) , Melanie Levesque (2) , Normand Amyot (2) , Claude Hudon (2) , Olivier Kokoko (2)

1 CEDRIC - LAETITIA - CEDRIC. Traitement du signal et architectures électroniques
2 IREQ - Institut de Recherche d'Hydro-Québec [Varennes]

Deep learning (DL) has been recently used in several applications of machine health monitoring systems. Unfortunately , most of these DL models are considered as black-boxes with low interpretability. In this research, we propose an original PHM framework based on visual data analysis. The most suitable space dimension for the data visualization is the 2D-space, which necessarily involves a significant reduction from a high-dimensional to a low-dimensional data space. To perform the data analysis and the diagnostic interpretation in a PHM framework, a Variational Autoencoder (VAE) is used jointly with a classifier. The proposed model was evaluated to automatically recognize individual Partial Discharge (PD) sources for hydro generators monitoring.

Mots clés: diagnosis analysis; Index Terms-Variational Autoencoder; data visualization; sys- tem health management

Domaines: Apprentissage [cs.LG]

Ryad Zemouri, Melanie Levesque, Normand Amyot, Claude Hudon, Olivier Kokoko. Deep Variational Autoencoder: An Efficient Tool for PHM Frameworks. 2020 Prognostics and Health Management Conference (PHM-Besançon), May 2020, Besancon, France. pp.235-240, ⟨10.1109/PHM-Besancon49106.2020.00046⟩. ⟨hal-02868384⟩