INVITED SPEAKER

NGUYEN Thi Phuong Khanh is currently an Associate Professor at National School of Engineering in Tarbes (ENIT), INP Toulouse, France, and member of the Production Engineering Laboratory since 2017.

She was co-chair of the invited track in the International Federation of Automatic Control World Congress from 2020 until now, technical program committee’s member of the European conference of PHM society 2020 and 2021, Doctoral Symposium in the European conference of PHM society 2022, and technical program committee’s co-chair of the European conference of PHM society 2024. She is involved in two research projects as Principal Investigator (Toulouse TECH InterLabs project in 2019 and ANR JCJC X-IMS project in 2022). Additionally, she is a co-scientific coordinator for numerous research projects, including one collaborative project with ALSTOM from 2017 to 2019, one regional project spanning from 2021 to 2024, and one European Interreg project running from 2017 to 2020. Furthermore, she regularly serves as an evaluator for academic articles for various journals such as IEEE, Elsevier, Springer, among others, as well as conferences including IEEE, IFAC, and PHM Society.

Title: Prediction of bearings remaining useful life based on contrastive self-supervised learning

Abstract:

This paper proposes a new contrastive self-supervised learning paradigm for bearing remaining useful life (RUL) prediction based on CNN-LSTM models. It addresses the dilemma of scarce labels and data imbalance in Prognostics and Health Management (PHM) by designing a specific pretext task to mine the potential degradation-related information in unlabelled data. In this paper, we propose a method to build contrastive sample pairs using sequence order information. Then, a Siamese CNN encoder guided by the customized contrastive loss is designed to maximize the differences between encoding features of the contrastive sample pairs. After that, the CNN’s parameters are partly frozen, and its encoded features are used as the input of the subsequent LSTM layer to predict the RUL. Finally, on the labeled dataset, LSTM is fine-tuned to optimize the ability of CNN-LSTM for RUL prediction. The proposed method is validated on “PRONOSTIA Bearing Dataset”. The obtained results and the analysis of the hidden layer output highlight the performance of the proposed approach, which outperforms the supervised learning paradigm in terms of maintaining the ability to capture sequential discriminatory information for better RUL prediction, especially in the case of a reduced amount of labeled data.