风力机轴承剩余寿命预测的解释性深度学习方法

余萍; 平梦梦; 马佳林; 曹洁

doi:doi:10.19912/j.0254-0096.tynxb.2024-0751

欢迎来到能源电力期刊网！

English Version 读者登录期刊登录

您当前的位置：

首页 >

文章列表页 >

风力机轴承剩余寿命预测的解释性深度学习方法

更新时间：2025-10-22

- 风力机轴承剩余寿命预测的解释性深度学习方法
- 暂无标题
- 太阳能学报 2025年46卷第9期页码：66-75
- 作者机构：
- 作者简介：
- 基金信息：
  
  国家自然科学基金（62241307）;甘肃省科技计划项目（22YF7FA166;23CXGA0060）;兰州市科技计划项目（2022-RC-60）
- DOI：doi:10.19912/j.0254-0096.tynxb.2024-0751
  中图分类号： TK83
- 网络出版：2025-10-14，
  
  纸质出版：2025
- 稿件说明：
移动端阅览
余萍, 平梦梦, 马佳林, 等. 风力机轴承剩余寿命预测的解释性深度学习方法[J]. 太阳能学报, 2025,46(9):66-75.
余萍, 平梦梦, 马佳林, 等. 风力机轴承剩余寿命预测的解释性深度学习方法[J]. 太阳能学报, 2025,46(9):66-75. DOI： doi:10.19912/j.0254-0096.tynxb.2024-0751.

DOI：

摘要

提出一种基于自注意力机制和位置编码的可解释性模型（PEARNN）

用于预测风力发电机轴承的剩余使用寿命。此模型结合了自注意力机制和位置编码的优势

通过分块掩码块和周期性位置编码来增强模型对序列数据的理解能力。实验部分在XJTUSY和IEEE PHM 2012这两个数据集上进行

验证了模型的有效性并可视化信号关注区域。结果表明

所提PEARNN模型在轴承故障预测任务中表现优异

具有良好的泛化性和鲁棒性。此外

该文还展示了模型通过可视化注意力机制如何关注不同的信号部分

从而提高预测的可信度和解释性。PEARNN模型不仅提高了故障预测的准确性

还增强了模型的可解释性。

Abstract

This study proposes an interpretable model

PEARNN

which integrates a self-attention mechanism and positional encoding for predicting the remaining useful life of bearings. The model combines the advantages of self-attention mechanism and positional encoding to enhance its understanding of sequential data through patch masking and periodic positional encoding. Experiments conducted on the XJTU-SY and IEEE PHM 2012 datasets validate the effectiveness of the model and visualize the attention regions of the signals. The results demonstrate that the proposed PEARNN model performs excellently in bearing fault prediction tasks

exhibiting good generalization and robustness. Additionally

the paper illustrates how the model identifies critical parts of the input signal through visualized attention mechanisms to improve prediction credibility and interpretability. PEARNN not only enhances the accuracy of fault prediction but also strengthens the interpretability of the model.

关键词

Keywords

references

GUO Y M, LIU Y, OERLEMANS A, et al.Deep learning for visual understanding: a review[J]. Neurocomputing, 2016, 187: 27-48.

ZHENG C, WU W H, CHEN C, et al.Deep learning-based human pose estimation: a survey[J]. ACM computing surveys, 2024, 56(1): 1-37.

CASTELVECCHI D.Can we open the black box of AI?[J]. Nature, 2016, 538(7623): 20-23.

余萍, 曹洁. 基于图形特征的双输入卷积神经网络风力机轴承剩余寿命预测[J]. 太阳能学报, 2022, 43(5): 343-350.

ZHAO C C, LIANG H Y, SHAN Y, et al.Bearing fault diagnosis method based on comprehensive information divergence and improved BP-AdaBoost algorithm[J]. Structural health monitoring, 2023, 22(5): 3047-3064.

XIANG C, REN Z J, SHI P F, et al.Data-driven fault diagnosis for rolling bearing based on DIT-FFT and XGBoost[J]. Complexity, 2021, 2021(1): 4941966.

FELDMANN C, BAJORATH J.Calculation of exact Shapley values for support vector machines with Tanimoto kernel enables model interpretation[J]. iScience, 2022, 25(9): 105023.

丁显, 徐进, 黎曦琳, 等. 融合维纳过程和粒子滤波的风力发电机轴承剩余寿命预测[J]. 太阳能学报, 2022, 43(12): 248-255.

钱小毅, 孙天贺, 王宝石, 等. 基于知识规则挖掘的风电机组故障识别与解释性分析[J]. 太阳能学报, 2023, 44(8): 379-385.

VASWANI A, SHAZEER N, PARMAR N, et al.Attention is all you need[C]//Proceedings of the 31st International Conference on Neural Information Processing Systems. Long Beach, California, USA, 2017: 6000-6010.

XIANG L, WANG P H, YANG X, et al.Fault detection of wind turbine based on SCADA data analysis using CNN and LSTM with attention mechanism[J]. Measurement, 2021, 175: 109094.

WANG H Q, LIN T J, CUI L L, et al.Multitask learning-based self-attention encoding atrous convolutional neural network for remaining useful life prediction[J]. IEEE transactions on instrumentation and measurement, 2022, 71: 3516108.

LIU C D, ZHANG L X, YAO R, et al.Dual attention-based temporal convolutional network for fault prognosis under time-varying operating conditions[J]. IEEE transactions on instrumentation and measurement, 2021, 70: 3512210.

谭启瑜, 马萍, 张宏立, 等. 基于图注意力网络的风力发电机齿轮箱故障诊断[J]. 太阳能学报, 2024, 45(1): 265-274.

LI A, XIAO F, ZHANG C, et al.Attention-based interpretable neural network for building cooling load prediction[J]. Applied energy, 2021, 299: 117238.

苏燕, 付家源, 林川, 等. 基于时间注意力机制的大坝动态变形预测模型[J]. 水力发电学报, 2022, 41(7): 72-84.

LIAO J X, DONG H C, SUN Z Q, et al.Attention-embedded quadratic network (qttention) for effective and interpretable bearing fault diagnosis[J]. IEEE transactions on instrumentation and measurement, 2023, 72: 3511113.

LI X, ZHANG W, DING Q.Understanding and improving deep learning-based rolling bearing fault diagnosis with attention mechanism[J]. Signal processing, 2019, 161: 136-154.

YANG S B, DENG Z G, LI X F, et al.A novel hybrid model based on STL decomposition and one-dimensional convolutional neural networks with positional encoding for significant wave height forecast[J]. Renewable energy, 2021, 173: 531-543.

邱大伟, 刘子辰, 周一青, 等. 基于Transformer神经网络的滚动轴承故障类型识别[J]. 高技术通讯, 2021, 31(1): 1-11.

孔雅洁, 张叶. 引入高斯掩码自注意力模块的YOLOv3目标检测方法[J]. 液晶与显示, 2022, 37(4): 539-548.

LI X, ZHANG W, DING Q.Deep learning-based remaining useful life estimation of bearings using multi-scale feature extraction[J]. Reliability engineering & system safety, 2019, 182: 208-218.

LI X, ZHANG W, MA H, et al.Data alignments in machinery remaining useful life prediction using deep adversarial neural networks[J]. Knowledge-based systems, 2020, 197: 105843.

SUH S, LUKOWICZ P, LEE Y O.Generalized multiscale feature extraction for remaining useful life prediction of bearings with generative adversarial networks[J]. Knowledge-based systems, 2022, 237: 107866.

XIA J, FENG Y W, LU C, et al.LSTM-based multi-layer self-attention method for remaining useful life estimation of mechanical systems[J]. Engineering failure analysis, 2021, 125: 105385.

ZHU J, CHEN N, PENG W W.Estimation of bearing remaining useful life based on multiscale convolutional neural network[J]. IEEE transactions on industrial electronics, 2019, 66(4): 3208-3216.

WANG B, LEI Y G, YAN T, et al.Recurrent convolutional neural network: a new framework for remaining useful life prediction of machinery[J]. Neurocomputing, 2020, 379: 117-129.

SU X Y, LIU H M, TAO L F, et al.An end-to-end framework for remaining useful life prediction of rolling bearing based on feature pre-extraction mechanism and deep adaptive transformer model[J]. Computers & industrial engineering, 2021, 161: 107531.

SHANG Y J, TANG X L, ZHAO G Q, et al.A remaining life prediction of rolling element bearings based on a bidirectional gate recurrent unit and convolution neural network[J]. Measurement, 2022, 202: 111893.

浏览量

下载量

CSCD

文章被引用时，请邮件提醒。

提交

工具集

关联资源

多度量下ResGAT的风力发电机齿轮箱故障诊断