Pre-Trained Large Language Model Based Remaining Useful Life Transfer Prediction of Bearing

TL;DR

This paper proposes a novel approach using pre-trained large language models to improve the transfer prediction of remaining useful life (RUL) for bearings, addressing data inconsistency and generalization issues in industrial settings.

Contribution

The study introduces a new method leveraging pre-trained language models for RUL prediction, enhancing transferability and robustness over traditional deep learning techniques.

Findings

Improved prediction accuracy in variable data conditions

Enhanced generalization for long-term RUL forecasting

Reduced impact of data distribution discrepancies

Abstract

Accurately predicting the remaining useful life (RUL) of rotating machinery, such as bearings, is essential for ensuring equipment reliability and minimizing unexpected industrial failures. Traditional data-driven deep learning methods face challenges in practical settings due to inconsistent training and testing data distributions and limited generalization for long-term predictions.