摘要详情

ID / 提交时间

21 / 2025-03-24 09:20:37

标题

Adaptive Remaining Useful Life Prediction Method for Devices With Random Failure Threshold

关键字

Remaining useful life (RUL),nonlinear,adaptive Wiener process,random failure threshold

主题及专题

S06 数据驱动的复杂装备故障诊断、预测及健康管理

状态

全文录用

作者

青董 / 火箭军工程大学

昌华胡 / 火箭军工程大学

党波杜 / 火箭军工程大学

迦陵张 / 火箭军工程大学

洪裴 / 火箭军工程大学

泽明张 / 火箭军工程大学

摘要

For nonlinear stochastic degraded devices in practical engineering, existing methods have overlooked the variability of adaptive drift as well as the impact of random failure threshold on remaining useful life (RUL) prediction. In this paper, we propose an adaptive RUL prediction method for nonlinear degraded devices with random failure threshold by utilizing the adaptive Wiener process. Firstly, a nonlinear degradation model incorporating adaptive drift is developed based on diffusion motion. The RUL life distribution, defined by the first hitting time, is derived while accounting for the impact of random failure threshold. Subsequently, the parameters of the degradation model and the distribution of random failure threshold are estimated offline based on historical monitoring data. The Kalman filter is utilized to dynamically update these estimation model parameters. Finally, the effectiveness of the proposed method is validated through case study of gyroscopes.