LT-Mamba: A Novel Network for RUL Estimation of Multisensor Signals

TY - JOUR

T1 - LT-Mamba

T2 - A Novel Network for RUL Estimation of Multisensor Signals

AU - Xie, Yining

AU - Lin, Xiaoyu

AU - Liu, Fuxiang

AU - Qi, Haochen

AU - Zhao, Jing

N1 - Publisher Copyright: © 2001-2012 IEEE.

PY - 2025

Y1 - 2025

N2 - Remaining useful life (RUL) prediction is a core task in the prognostics and health management (PHM) of mechanical systems, which helps to guide preventive maintenance to improve the reliability of industrial systems. However, current models struggle to effectively model the “implicit operating condition” changes caused by the coupling of degradation in multiple components when faced with complex machinery. Therefore, in this article, we propose a multichannel integrated network based on state-space models, called the LT-Mamba network, which models operating conditions while providing deep interpretability. Specifically, we introduce a Mamba model based on state-space models (SSMs) to enhance the model’s perception and fitting capabilities for “implicit operating condition” variations. Additionally, we augment Mamba’s global observation capabilities with a trend auxiliary attention mechanism (TAAM) and LSTM to increase sensitivity to changes in degradation rates. Finally, we conduct life prediction by integrating features extracted from various models. We conducted benchmark experiments using the C-MAPSS dataset and the Milling dataset. The results demonstrate that the proposed method outperforms the existing methods.

AB - Remaining useful life (RUL) prediction is a core task in the prognostics and health management (PHM) of mechanical systems, which helps to guide preventive maintenance to improve the reliability of industrial systems. However, current models struggle to effectively model the “implicit operating condition” changes caused by the coupling of degradation in multiple components when faced with complex machinery. Therefore, in this article, we propose a multichannel integrated network based on state-space models, called the LT-Mamba network, which models operating conditions while providing deep interpretability. Specifically, we introduce a Mamba model based on state-space models (SSMs) to enhance the model’s perception and fitting capabilities for “implicit operating condition” variations. Additionally, we augment Mamba’s global observation capabilities with a trend auxiliary attention mechanism (TAAM) and LSTM to increase sensitivity to changes in degradation rates. Finally, we conduct life prediction by integrating features extracted from various models. We conducted benchmark experiments using the C-MAPSS dataset and the Milling dataset. The results demonstrate that the proposed method outperforms the existing methods.

KW - Deep learning

KW - Mamba

KW - multisensor signals

KW - prognostics and health management (PHM)

KW - remaining useful life (RUL) estimation

UR - https://www.scopus.com/pages/publications/105016501412

U2 - 10.1109/JSEN.2025.3608799

DO - 10.1109/JSEN.2025.3608799

M3 - Article

AN - SCOPUS:105016501412

SN - 1530-437X

VL - 25

SP - 40450

EP - 40460

JO - IEEE Sensors Journal

JF - IEEE Sensors Journal

IS - 21

ER -