[1]王 骅,周俊宏,李文泽,等.基于SSA-VMD-CNN-BiLSTM 的张力机轴承故障分级诊断[J].机械与电子,2026,44(08):26-34.
　WANG Hua,ZHOU Junhong,LI Wenze,et al.Multi-level Fault Diagnosis of Tensioner Bearings Based on SSA-VMD-CNN-BiLSTM[J].Machinery & Electronics,2026,44(08):26-34.

点击复制

基于SSA-VMD-CNN-BiLSTM 的张力机轴承故障分级诊断()

分享到：

《机械与电子》[ISSN:1001-2257/CN:52-1052/TH]

卷:

44

期数:

2026年08期

页码:

26-34

栏目:

智能检测

出版日期:

2026-08-25

文章信息/Info

Title:

Multi-level Fault Diagnosis of Tensioner Bearings Based on SSA-VMD-CNN-BiLSTM

文章编号:

1001-2257(2026)08-0026-09

作者:

王 骅¹; 周俊宏¹; 李文泽¹; 王海燕¹; 吴伟智²; 杨 先²

1.广东电网有限责任公司惠州供电局,广东 惠州 516001;

2.广东电网能源发展有限公司,广东 广州 510160

Author(s):

WANG Hua¹; ZHOU Junhong¹; LI Wenze¹; WANG Haiyan¹; WU Weizhi²; YANG Xian²

(1.Huizhou Power Supply Bureau,Guangdong Power Grid Co.,Ltd.,Huizhou 516001,China;

2.Guangdong Power Grid Energy Development Co.,Ltd.,Guangzhou 510160,China)

关键词:

张力机; 轴承; 故障诊断; 麻雀搜索算法; 双向长短期记忆网络

Keywords:

tensioner; bearing; fault diagnosis; sparrow search algorithm; bidirectional long short-term memory network

分类号:

TH133.3;TP183

文献标志码:

A

摘要:

针对新能源张力机轴承在低速重载、频繁启停和张力波动工况下故障特征微弱、非平稳性强以及多级分类识别困难的问题,构建了一种融合麻雀搜索算法(SSA)优化变分模态分解(VMD)、卷积神经网络(CNN)与双向长短时记忆网络(BiLSTM)的SSA VMD CNN BiLSTM 轴承故障诊断模型。该模型以本征模态分量(IMF)排列熵之和作为适应度函数,引入麻雀搜索算法对VMD模态数K 和惩罚因子α 进行自适应优化,以获得时频聚集性和故障敏感性更优的本征模态分量。随后,将优化后的IMF分量输入CNN提取局部深层特征,并通过BiLSTM 对故障特征序列的双向时序相关性进行建模,实现新能源张力机发电机主轴轴承正常、内圈故障、外圈故障和滚动体故障等状态的分类识别。实验结果表明,排列熵作为SSA 寻优适应度函数时模型诊断性能较优,所提出的SSA-VMD CNN-BiLSTM 模型测试集分类准确率达到98.33%,Macro F1值达到98.31%,平均运行时间为11.04 s,均优于CNN-BiLSTM 和SSA-VMD-CNN-LSTM 等对比模型。多指标评价、混淆矩阵、统计显著性检验和消融实验结果进一步表明,SSA-VMD、CNN 和BiLSTM 各模块能够分别增强故障敏感分量表达、局部特征提取和双向时序建模能力。

Abstract:

To address the challenges of the weak fault signatures,strong non-stationarity,and difficulty in multi-class fault classification for bearings in new-energy tensioners operating under low-speed heavy-load,frequent start-stop,and tension fluctuation conditions,this paper proposes a hybrid fault diagnosis model that integrates Sparrow Search Algorithm (SSA)-optimized Variational Mode Decomposition (VMD),Convolutional Neural Network (CNN),and Bidirectional Long Short-Term Memory (BiLSTM), denoted as SSA – VMD– CNN –BiLSTM.In the proposed model,the sum of permutation entropy of the intrinsic mode functions (IMFs) is adopted as the fitness function,and the SSA is introduced to adaptively optimize the number of decomposition modes K and penalty factor α in VMD,thereby yielding IMFs with superior time-frequency concentration and fault sensitivity.The optimized IMF components are then fed into a CNN to extract local deep features,followed by a BiLSTM network to model the bidirecional temporal dependencies of fault feature sequences,enabling the classification of four states of the generator spindle bearing in new energy tensioners:normal state,inner race fault,outer race fault,and rolling element fault.Experimental results show that using permutation entropy as the SSA fitness function yields the best diagnostic performance.The proposed model achieves a test-set classification accuracy of 98.33%,a Macro-F1 value of 98.31%,and an average running time of 11.04 s,outperforming comparison models,including the CNN-BiLSTM and SSA-VMD-CNN-LSTM.Multi-metricevaluation,confusion matrix analysis,statistical significance tests,and ablation experiments further demonstrate that the SSA-VMD,CNN,and BiLSTM modules respectively enhance fault sensitive component representation,local feature extraction,and bidirectional temporal modeling capability.

备注/Memo

备注/Memo:

收稿日期:2026-04-28

基金项目:广东电网有限责任公司科技项目(031300KC23120022(GDKJXM20231447))

作者简介:王 骅 (1990-),男,高级工程师,研究方向为电气工程及电力基建等。

常用功能