退化趋势平滑约束下基于BLSTM-VAE的剩余寿命预测方法

王旋; 石章松; 佘博; 孙世岩; 秦奋起

doi:10.12382/bgxb.2024.0635

您当前的位置：

首页 >

文章列表页 >

退化趋势平滑约束下基于BLSTM-VAE的剩余寿命预测方法

更新时间：2025-09-05

- 退化趋势平滑约束下基于BLSTM-VAE的剩余寿命预测方法
- Remaining Life Prediction Method Based on BLSTMN-VAE under Degradation Trend Smoothing Constraint
- 兵工学报 2025年46卷第5期页码：240635
- 作者机构：
  
  1. 海军工程大学, 湖北武汉 430033
  2. 中国船舶重工集团公司第七一三研究所, 河南郑州 450052
- 作者简介：
  
  * 邮箱: 1802303@nue.edu.cn
- 基金信息：
  
  国家自然科学基金项目(61640308)
- DOI：10.12382/bgxb.2024.0635
  中图分类号：
- 收稿：2024-07-26，
  
  网络出版：2025-05-07，
  
  纸质出版：2025-05-31
- 稿件说明：
移动端阅览
王旋, 石章松, 佘博, 等. 退化趋势平滑约束下基于BLSTM-VAE的剩余寿命预测方法[J]. 兵工学报, 2025,46(5):240635.

Xuan WANG, Zhangsong SHI, Bo SHE, et al. Remaining Life Prediction Method Based on BLSTMN-VAE under Degradation Trend Smoothing Constraint[J]. Acta Armamentarii, 2025, 46(5): 240635.
王旋, 石章松, 佘博, 等. 退化趋势平滑约束下基于BLSTM-VAE的剩余寿命预测方法[J]. 兵工学报, 2025,46(5):240635. DOI： 10.12382/bgxb.2024.0635.

Xuan WANG, Zhangsong SHI, Bo SHE, et al. Remaining Life Prediction Method Based on BLSTMN-VAE under Degradation Trend Smoothing Constraint[J]. Acta Armamentarii, 2025, 46(5): 240635. DOI： 10.12382/bgxb.2024.0635.

摘要

剩余寿命(Remaining Useful Life

RUL)预测对于维护工业设备的可靠性和安全性至关重要

但现有的RUL预测方法在处理高维传感器数据以及捕捉时间退化模式方面仍然面临诸多挑战。为了解决上述问题

提出一种退化趋势平滑约束下基于双向长短时记忆网络-变分自编码器(Bidirectional Long Short Term-Memory-Variational Auto Encoder

BLSTM-VAE)的RUL预测方法。该方法首先进行数据预处理

包括数据降噪、滑动窗口分段和标签修正等步骤。然后设计基于BLSTM的VAE型特征提取器

以有效提取时间序列数据中的非线性关系和长距离依赖关系。最后提出一种基于流形学习的退化趋势平滑约束模块

通过局部不变性假设来增强模型的稳健性和泛化能力。通过航空发动机数据集数据集进行验证

结果表明所提出的RUL预测方法在数据集上的表现优于现有的多种RUL预测方法

具有更低的预测误差和更高的稳定性。

Abstract

Remaining useful life (RUL) prediction is crucial for maintaining the reliability and safety of industrial equipment

but the existing RUL prediction methods still face many challenges in processing the high-dimensional sensor data and capturing the temporal degradation patterns. To address the above issues

this paper proposes a RUL prediction method based on bidirectional long short term memory network-variational auto encoder (BLSTMN-VAE) under the constraint of degradation trend smoothing. This method is used for data preprocessing

including data noise reduction

sliding window segmentation

and label correction. Then

a BLSTMN-based VAE type feature extractor is designed to effectively extract the nonlinear relationships and long-distance dependencies in time series data. Finally

a degradation trend smoothing constraint module based on manifold learning is proposed to enhance the robustness and generalization ability of the proposed model through the assumption of local invariance. The proposed RUL prediction method is verified using the aero-engine dataset. The results show that the proposed RUL prediction method outperforms various existing RUL prediction methods

and has lower prediction errors and higher stability.

关键词

Keywords

references

陈子涵 . 基于多模态Transformer的机电作动器剩余寿命预测 [J ] . 兵工学报 , 2023 , 44 ( 10 ): 2920 - 2931 . DOI: 10.12382/bgxb.2022.0581 http://doi.org/10.12382/bgxb.2022.0581 机电作动器在航空航天装备中扮演着重要角色。针对机电作动器剩余寿命预测问题,提出一种基于多模态Transformer模型的机电作动器寿命预测方法。该方法直接使用多通道传感器数据作为输入,综合考虑多模态数据信息,并且不需要人工特征提取等预处理步骤。多模态Transformer模型利用多头自注意力机制从不同的表示子空间中自适应学习全局特征,能够避免传统深度学习方法难以学习全局特征的缺点。利用多模态Transformer的编码器部分并行提取多模态传感器时间序列中不同传感器的特征,并实时直接预测剩余使用寿命。采用由编码器和解码器组成的完整多模态Transformer模型预测机电作动器的关键性能参数,可同时更直观地预测关键寿命参数的退化过程。使用机电作动器全寿命试验数据验证该方法用于寿命预测的有效性。试验结果表明,所提方法能够准确地直接预测剩余寿命,同时预测关键性能参数的寿命退化过程。

CHEN Z H . Prognosticating remaining useful life of electro-mechanical actuators using a multi-mode transformermodel [J ] . Acta Armamentarii , 2023 , 44 ( 10 ): 2920 - 2931 . (in Chinese)

WANG B , LEI Y G , LI N P , et al. Multiscale convolutional attention network for predicting remaining useful life of machinery [J ] . IEEE Transactions on Industrial Electronics , 2021 , 68 ( 8 ): 7496 - 7504 .

CAO Y D , DING Y F , JIA M P , et al. A novel temporal convolutional network with residual self-attention mechanism for remaining useful life prediction of rolling bearings [J ] . Reliability Engineering & System Safety , 2021 , 215 : 107813 .

许晓东 , 唐圣金 , 谢建 , 等 . 随机退化应力作用下设备剩余寿命预测方法 [J ] . 兵工学报 , 2022 , 43 ( 3 ): 712 - 719 . DOI: 10.12382/bgxb.2021.0018 http://doi.org/10.12382/bgxb.2021.0018 有效的剩余寿命预测对设备可靠性与安全性起着关键作用，不确定的内部老化状态和外部使用工况导致的随机退化应力会极大地影响设备的退化速率和健康状况。提出一种新的随机退化应力作用下的设备剩余寿命预测方法。将随机退化应力引入老化过程，基于维纳过程建立随机应力作用下的设备老化模型，提出融合期望最大化算法和粒子群优化算法的先验参数离线估计方法；在贝叶斯框架下在线更新随机系数，推导剩余寿命预测结果的概率密度分布函数。基于锂电池实验退化数据验证了所提方法的有效性。结果表明，考虑随机应力对设备退化规律的影响，能够有效提高剩余寿命的预测精度并降低预测结果的不确定性。

XU X D , TANG S J , XIE J , et al. Remaining useful life prediction of equipment under random degradation stress [J ] . Acta Armamentarii , 2022 , 43 ( 3 ): 712 - 719 . (in Chinese) DOI: 10.12382/bgxb.2021.0018 http://doi.org/10.12382/bgxb.2021.0018 The effective remaining useful life prediction plays a key role in improving the reliability and safety of equipment. The uncertain internal aging state and external working condition have great affect on the degradation rate and state-of-health of equipment. A novel method for predicting the remaining useful life of equipment under the random stress is proposed. The random degradation stress is introduced into the aging process of equipment，and a degradation model for equipment is established based on Wiener process. An off-line prior parameter estimation method based on expectation maximization algorithm and particle swarm optimization algorithm is proposed. The random parameter is updated online in Bayesian framework，and the probability distribution function of the remaining useful life prediction result is derived. The proposed method is verified by the experimental degradation data of lithium-ion batteries. The results show that the proposed method can be used to effectively improve the prediction accuracy of remaining useful life and reduce the uncertainty of prediction results in considering the influence of random stress on the degradation law of equipment.

刘小平 , 张立杰 , 沈凯凯 , 等 . 考虑测量误差的步进加速退化试验建模与剩余寿命估计 [J ] . 兵工学报 , 2017 , 38 ( 8 ): 1586 - 1592 . DOI: 10.3969/j.issn.1000-1093.2017.08.017 http://doi.org/10.3969/j.issn.1000-1093.2017.08.017 步进应力加速退化试验已发展成高可靠、长寿命产品可靠性评估与剩余寿命估计的主要试验方法。为研究测量误差在基于步进应力加速退化试验方法剩余寿命估计中的影响，建立了基于Wiener过程的考虑个体差异和测量误差的退化模型。将Wiener过程的漂移系数随机化描述个体差异，在首达时间意义下得到了寿命分布的概率密度函数。基于极大似然估计法对模型中引入的未知参数进行估计。采用蒙特卡洛方法对激光器的性能退化进行了仿真研究。研究结果表明，考虑测量误差的退化模型的模型拟合性和剩余寿命估计精度都优于不考虑测量误差的方法，可以提高可靠性估计的精度与剩余寿命预测的准确性。

LIU X P , ZHANG L J , SHEN K K , et al. Step stress accelerated degradation test modeling and remaining useful life estimation in consideration of measuring error [J ] . Acta Armamentarii , 2017 , 38 ( 8 ): 1586 - 1592 . (in Chinese)

JOUIN M , GOURIVEAU R , HISSEL D , et al. Degradations analysis and aging modeling for health assessment and prognostics of PEMFC [J ] . Reliability Engineering & System Safety , 2016 , 148 : 78 - 95 .

TIAN H X , REN D X , LI K , et al. An adaptive update model based on improved long short term memory for online prediction of vibration signal [J ] . Journal of Intelligent Manufacturing , 2021 , 32 ( 1 ): 37 - 49 .

ZHAO R , WANG D Z , YAN R Q , et al. Machine health monitoring using local feature-based gated recurrent unit networks [J ] . IEEE Transactions on Industrial Electronics , 2018 , 65 ( 2 ): 1539 - 1548 .

梁伟阁 , 闫啸家 , 佘博 , 等 . 基于FA-LN-BiGRU的机械设备剩余寿命区间预测方法 [J ] . 振动.测试与诊断 , 2023 , 43 ( 3 ): 513 - 519 ,620-621.

LIANG W G , YAN X J , SHE B , et al. Remaining useful life interval prediction of mechanical equipment based on FA-LN-BiGRU [J ] . Journal of Vibration, Measurement & Diagnosis , 2023 , 43 ( 3 ):513-519, 620 - 621 . (in Chinese)

朱挺 , 陈兆祥 , 周笛 , 等 . 基于Bayesian-LSTM神经网络的热轧轧辊剩余寿命预测及不确定性评估 [J ] . 机械工程学报 , 2024 , 60 ( 11 ): 181 - 190 .

ZHU T , CHEN Z X , ZHOU D , et al. Bayesian-LSTM neural network-based remaining useful life prediction and uncertainty estimation of rollers in a hot strip Mill [J ] . Journal of Mechanical Engineering , 2024 , 60 ( 11 ): 181 - 190 . (in Chinese)

YU W N , KIM I Y , MECHEFSKE C . Analysis of different RNN autoencoder variants for time series classification and machine prognostics [J ] . Mechanical Systems and Signal Processing , 2021 , 149 : 107322 .

张鲁一航 , 杨彦明 , 陈永展 , 等 . 基于VMD-CNN-BiLSTM的变工况涡扇发动机剩余寿命预测 [J ] . 北京航空航天大学学报 . https://doi.org/10.13700/j.bh.1001-5965.2021.0051. https://doi.org/10.13700/j.bh.1001-5965.2021.0051. https://doi.org/10.13700/j.bh.1001-5965.2021.0051. https://doi.org/10.13700/j.bh.1001-5965.2021.0051.

ZHANG L Y H , YANG Y M , CHEN Y Z , et al. Remaining useful life life prediction of variable-operating turbofan engine based on VMD-CNN-BiLSTM [J ] . Journal of Beijing University of Aeronautics and Astronautics . https://doi.org/10.13700/j.bh.1001-5965.2021.0051. https://doi.org/10.13700/j.bh.1001-5965.2021.0051. (in Chinese) https://doi.org/10.13700/j.bh.1001-5965.2021.0051. https://doi.org/10.13700/j.bh.1001-5965.2021.0051.

YU W N , KIM I Y , MECHEFSKE C . Remaining useful life estimation using a bidirectional recurrent neural network based autoencoder scheme [J ] . Mechanical Systems and Signal Processing , 2019 , 129 : 764 - 780 .

CHENG Y W , WANG C , WU J , et al. Multi-dimensional recurrent neural network for remaining useful life prediction under variable operating conditions and multiple fault modes [J ] . IEEE Sensors Journal , 2021 , 21 ( 9 ): 10905 - 10914 .

COSTA N , SANCHEZ L . Variational encoding approach for interpretable assessment of remaining useful life estimation [J ] . Reliability Engineering & System Safety , 2022 , 222 : 108353 .

BEKTAS O . A neural network filtering approach for similarity-based remaining useful life estimation [J ] . The International Journal of Advanced Manufacturing Technology , 2019 ,101: 87 - 103 .

XU X W , LI X , MING W W , et al. A novel multi-scale CNN and attention mechanism method with multi-sensor signal for remaining useful life prediction [J ] . Computers & Industrial Engineering , 2022 , 169 : 108 - 204 .

BELKIN M , NIYOGI P , SINDHWANI V . Manifold regularization: a geometric framework for learning from labeled and unlabeled examples [J ] . Journal of Machine Learning Research , 2006 ( 7 ): 2399 - 2434 .

ROWEIS S T , SAUL L K . Nonlinear dimensionality reduction by locally linear embedding [J ] . Science , 2000 , 290 ( 5500 ): 2323 - 2326 . DOI: 10.1126/science.290.5500.2323 http://doi.org/10.1126/science.290.5500.2323 Many areas of science depend on exploratory data analysis and visualization. The need to analyze large amounts of multivariate data raises the fundamental problem of dimensionality reduction: how to discover compact representations of high-dimensional data. Here, we introduce locally linear embedding (LLE), an unsupervised learning algorithm that computes low-dimensional, neighborhood-preserving embeddings of high-dimensional inputs. Unlike clustering methods for local dimensionality reduction, LLE maps its inputs into a single global coordinate system of lower dimensionality, and its optimizations do not involve local minima. By exploiting the local symmetries of linear reconstructions, LLE is able to learn the global structure of nonlinear manifolds, such as those generated by images of faces or documents of text.

SAXENA A , GOEBEL K , SIMON D , et al. Damage propagation modeling for aircraft engine run-to-failure simulation [C ] // Proceedings of 2008 International Conference on Prognostics and Health Management. Denver, CO , US : IEEE , 2008 : 1 - 9 .

WANG Z J , ZHAO W Y , LI Y F , et al. Adaptive staged RUL prediction of rolling bearing [J ] . Measurement , 2023 , 222 : 113478 .

ZHENG S , RISTOVSKI K , FARAHAT A , et al. Long short-term memory network for remaining useful life estimation [C ] // Proceedings of 2017 IEEE International Conference on Prognostics and Health Management. Dallas, TX , US : IEEE , 2017 : 88 - 95 .

WANG T C . Remaining useful life predictions for turbofan engine degradation based on concurrent semi-supervised model [J ] . Neural Computing and Applications , 2022 ,34: 5151 - 5160 .

LI H , ZHAO W , ZHANG Y X , et al. Remaining useful life prediction using multi-scale deep convolutional neural network [J ] . Applied Soft Computing , 2020 , 89 : 106113 .

CHENG Y W , HU K , WU J , et al. Autoencoder quasi-recurrent neural networks for remaining useful life prediction of engineering systems [J ] . IEEE/ASME Transactions on Mechatronics , 2022 , 27 ( 2 ): 1081 - 1092 .

浏览量

220

下载量

CNKI被引量

文章被引用时，请邮件提醒。

提交

工具集

关联资源

基于改进测地距离的等度规映射及其在声目标特征提取中的应用

Complex structure of human Hsp90^N and a novel small inhibitor FS5

Recent studies on potential accident-tolerant fuel-cladding systems in light water reactors

Monte Carlo simulation of lattice analysis of complex LaCeTh_0.1CuO_y using ion bombardment technique

Spin coating of TPB film on acrylic substrate and measurement of its wavelength shifting efficiency