基于ASFF-BiGRU的滚动轴承剩余寿命预测方法研究  

Research on Residual Life Prediction Method of Rolling BearingsBased on ASFF-BiGRU

作　　者：王誉霏 于洋[1] WANG Yufei;YU Yang(School of Information Science and Engineering,Shenyang University of Technology,Shenyang 110870,China)

机构地区：[1]沈阳工业大学信息科学与工程学院,沈阳110870

出　　处：《微处理机》2025年第2期61-64,共4页Microprocessors

摘　　要：针对滚动轴承剩余使用寿命预测中数据驱动方法难以有效提取敏感特征信息的问题,提出一种基于ASFF-BiGRU的预测方法。该方法对滚动轴承的声发射信号进行时、频域特征提取,结合单调性和时间相关性筛选敏感特征集,并通过自适应特征融合策略构建健康指标。利用双向门控循环单元(BiGRU)对健康指标进行趋势预测,实现RUL的精准预测。实验结果表明,与经典门控循环单元网络方法相比,所提方法有效提高了轴承剩余寿命的预测精度,为滚动轴承的健康管理提供了更可靠的技术支持。Aiming at the problem that data-driven methods struggle to effectively extract sensitive feature information in the prediction of the Remaining Useful Life of rolling bearings,a prediction method based on ASFF-BiGRU is proposed.This method extracts time-domain and frequency-domain features from the acoustic emission signals of rolling bearings,selects sensitive feature sets by combining monoto-nicity and temporal correlation,and constructs health indicators using an Adaptive Spatial Feature Fusion strategy.The Bidirectional Gated Recurrent Unit is then utilized to predict the trend of the health indicators,achieving accurate RUL prediction.Experimental results show that,compared to the classical gated re-current unit network method,the proposed method significantly improves the prediction accuracy of bearing remaining life,providing more reliable technical support for the health management of rolling bearings.

关 键 词：滚动轴承 剩余寿命预测 自适应特征融合 双向门控循环单元 

分 类 号：TH133.33[机械工程—机械制造及自动化]