蜉蝣优化双通道网络在齿轮箱故障诊断中的应用

于宁; 魏沉潜; 田立勇; 赵建军; 于晓涵

doi:10.7652/xjtuxb202505021

您当前的位置：

首页 >

文章列表页 >

蜉蝣优化双通道网络在齿轮箱故障诊断中的应用

—————————————————————— | 更新时间：2025-08-26

- 蜉蝣优化双通道网络在齿轮箱故障诊断中的应用
- Application of Mayfly Algorithm Optimized Dual-Channel Network in Gearbox Fault Diagnosis
- 西安交通大学学报 2025年59卷第5期页码：217-228
- 作者机构：
  
  辽宁工程技术大学机械工程学院， 123000，辽宁阜新
- 作者简介：
  
  于宁（1979—），女，教授，硕士生导师。
- 基金信息：
  
  国家自然科学基金资助项目(52174143)
- DOI：10.7652/xjtuxb202505021
  中图分类号： TH133.3
- 收稿：2024-09-21，
  
  网络首发：2025-01-10，
  
  纸质出版：2025-05-10
- 稿件说明：
移动端阅览
于宁, 魏沉潜, 田立勇, 等. 蜉蝣优化双通道网络在齿轮箱故障诊断中的应用[J]. 西安交通大学学报, 2025,59(5):217-228.

YU Ning, WEI Chenqian, TIAN Liyong, et al. Application of Mayfly Algorithm Optimized Dual-Channel Network in Gearbox Fault Diagnosis[J]. Journal of Xi’an Jiaotong University, 2025, 59(5): 217-228.
于宁, 魏沉潜, 田立勇, 等. 蜉蝣优化双通道网络在齿轮箱故障诊断中的应用[J]. 西安交通大学学报, 2025,59(5):217-228. DOI： 10.7652/xjtuxb202505021.

YU Ning, WEI Chenqian, TIAN Liyong, et al. Application of Mayfly Algorithm Optimized Dual-Channel Network in Gearbox Fault Diagnosis[J]. Journal of Xi’an Jiaotong University, 2025, 59(5): 217-228. DOI： 10.7652/xjtuxb202505021.

摘要

为了有效提取齿轮箱中齿轮和滚动轴承的故障信号特征，并克服深度学习模型超参数选取依赖人工经验的局限性，提高故障诊断的准确性和稳定性，提出了一种基于蜉蝣算法（MA）优化的双通道神经网络故障诊断模型。该模型采用一维时序输入的门控循环单元（GRU）和二维图像输入的卷积神经网络（CNN）构建双通道并行架构，并引入自适应批标准化（AdaBN）算法。利用MA的全局优化能力，以CNN-GRU的诊断精度为优化目标，自适应调整模型超参数。将蜉蝣算法优化效果与粒子群算法和遗传算法进行了对比验证，以评估其在模型参数优化方面的有效性。基于东南大学齿轮箱数据集和凯斯西储大学轴承数据集的实验结果表明：该模型能够有效提取振动信号特征，其故障识别精度与稳定性均优于典型深度学习模型，并展现出较强的鲁棒性。在稳态工况下，优化后的CNN-GRU（MA-CNN-GRU）模型在各数据集上的识别精度显著提高；在噪声工况下，MA优化的CNN-GRU模型表现出优异的抗噪性；在变负载工况下，结合AdaBN算法的MA-CNN-GRU模型实现了最高的平均识别精度。所提模型能够高效、准确地检测齿轮箱故障，为机械设备的维护和稳定运行提供了重要的参考价值。

Abstract

In order to effectively extract the fault signal characteristics of gears and rolling bearings in gearboxes

overcome the limitation of deep learning model hyperparameter selection relying on human experience

and improve the accuracy and stability of fault diagnosis

a two-channel neural network fault diagnosis model based on Mayfly algorithm (MA) optimization was proposed. The model uses a one-dimensional timing input gated recurrent unit (GRU) and a two-dimensional image input convolutional neural network (CNN) to construct a dual-channel parallel architecture

and introduces an adaptive batch normalization (AdaBN) algorithm. By using the global optimization ability of MA

the hyperparameters of the model are adaptively adjusted with the diagnostic accuracy of CNN-GRU as the optimization objective. At the same time

MA optimization effort is compared with particle swarm optimization and genetic algorithm to evaluate its effectiveness in model parameter optimization. The experimental results based on the gearbox data set of Southeast University and the bearing data set of Case Western Reserve University show that the model can effectively extract the characteristics of vibration signals

and its fault recognition accuracy and stability are better than typical deep learning models

and it shows strong robustness. Under steady-state conditions

the recognition accuracy of the optimized CNN-GRU model on each data set is significantly improved. Under noise conditions

the MA-optimized CNN-GRU (MA-CNN-GRU) model exhibits excellent noise immunity. Under variable load conditions

the MA-CNN-GRU model combined with the AdaBN algorithm achieves the highest average recognition accuracy. In summary

the proposed model can detect gearbox faults efficiently and accurately

and it can provide an important reference for the maintenance and stable operation of mechanical equipment.

关键词

Keywords

references

陈向民 , 舒文伊 , 韩梦茹 , 等 . 基于URP-ANCNN的变转速齿轮箱智能故障诊断方法 [J ] . 噪声与振动控制 , 2024 , 44 ( 2 ): 129 - 135 .

CHEN Xiangmin , SHU Wenyi , HAN Mengru , et al . Intelligence fault diagnosis method for gearboxes under variable rotational speed based on URP-ANCNN [J ] . Noise and Vibration Control , 2024 , 44 ( 2 ): 129 - 135 .

谢锋云 , 汪淦 , 王玲岚 , 等 . STFT结合2D CNN-SVM的齿轮箱故障诊断方法 [J ] . 噪声与振动控制 , 2024 , 44 ( 4 ): 103 - 109 .

XIE Fengyun , WANG Gan , WANG Linglan , et al . Fault diagnosis method of gearbox based on STFT and 2DCNN-SVM [J ] . Noise and Vibration Control , 2024 , 44 ( 4 ): 103 - 109 .

吴胜利 , 周燚 , 邢文婷 . 基于SDP和MCNN-LSTM的齿轮箱故障诊断方法 [J ] . 振动与冲击 , 2024 , 43 ( 15 ): 126 - 132 .

WU Shengli , ZHOU Yi , XING Wenting . Gearbox fault diagnosis based on SDP and MCNN-LSTM [J ] . Journal of Vibration and Shock , 2024 , 43 ( 15 ): 126 - 132 .

陈子旭 , 朱振杰 , 卢国梁 . 一种新的图谱域滚动轴承早期故障检测与识别方法 [J ] . 振动与冲击 , 2022 , 41 ( 6 ): 51 - 59 .

CHEN Zixu , ZHU Zhenjie , LU Guoliang . Novel early fault detection and diagnosis for rolling element bearings in graph spectrum domain [J ] . Journal of Vibration and Shock , 2022 , 41 ( 6 ): 51 - 59 .

王岩红 , 温笑欢 , 揭永琴 , 等 . 基于对比学习的滚动轴承早期故障在线检测方法 [J ] . 振动与冲击 , 2023 , 42 ( 14 ): 229 - 236 .

WANG Yanhong , WEN Xiaohuan , JIE Yongqin , et al . Online detection method for bearing incipient faults based on contrastive learning [J ] . Journal of Vibration and Shock , 2023 , 42 ( 14 ): 229 - 236 .

李迎秋 , 周威 , 郝德成 . 基于孪生网络的旋转机械故障诊断方法 [J ] . 组合机床与自动化加工技术 , 2023 ( 11 ): 116 - 120 .

LI Yingqiu , ZHOU Wei , HAO Decheng . A fault diagnosis method for rotating machinery based on twin networks [J ] . Modular Machine Tool & Automatic Manufacturing Technique , 2023 ( 11 ): 116 - 120 .

ZHANG Jing , TIAN Jing , WEN Tao , et al . Deep fault diagnosis for rotating machinery with scarce labeled samples [J ] . Chinese Journal of Electronics , 2020 , 29 ( 4 ): 693 - 704 .

王椿晶 , 王海瑞 . 深度在线小波极限学习在旋转机械故障诊断中的应用 [J ] . 机械科学与技术 , 2023 , 42 ( 7 ): 1029 - 1034 .

WANG Chunjing , WANG Hairui . Application of depth online wavelet extreme learning machine in rotating machinery fault diagnosis [J ] . Mechanical Science and Technology for Aerospace Engineering , 2023 , 42 ( 7 ): 1029 - 1034 .

吴春志 , 冯辅周 , 吴守军 , 等 . 深度学习在旋转机械设备故障诊断中的应用研究综述 [J ] . 噪声与振动控制 , 2019 , 39 ( 5 ): 1 - 7 .

WU Chunzhi , FENG Fuzhou , WU Shoujun , et al . Review of application research of deep learning in fault diagnosis of rotating machinery [J ] . Noise and Vibration Control , 2019 , 39 ( 5 ): 1 - 7 .

李益兵 , 马建波 , 江丽 . 基于SFLA改进卷积神经网络的滚动轴承故障诊断 [J ] . 振动与冲击 , 2020 , 39 ( 24 ): 187 - 193 .

LI Yibing , MA Jianbo , JIANG Li . Fault diagnosis of rolling bearing based on an improved convolutional neural network using SFLA [J ] . Journal of Vibration and Shock , 2020 , 39 ( 24 ): 187 - 193 .

陈功胜 , 唐向红 , 陆见光 , 等 . 基于CNN-ETR的滚动轴承故障诊断研究 [J ] . 兵器装备工程学报 , 2021 , 42 ( 6 ): 251 - 255 .

CHEN Gongsheng , TANG Xianghong , LU Jianguang , et al . Research on fault diagnosis of rolling bearing based on CNN-ETR [J ] . Journal of Ordnance Equipment Engineering , 2021 , 42 ( 6 ): 251 - 255 .

赵小强 , 张亚洲 . 利用改进卷积神经网络的滚动轴承变工况故障诊断方法 [J ] . 西安交通大学学报 , 2021 , 55 ( 12 ): 108 - 118 .

ZHAO Xiaoqiang , ZHANG Yazhou . Improved CNN-based fault diagnosis method for rolling bearings under variable working conditions [J ] . Journal of Xi'an Jiaotong University , 2021 , 55 ( 12 ): 108 - 118 .

肖雄 , 王健翔 , 张勇军 , 等 . 一种用于轴承故障诊断的二维卷积神经网络优化方法 [J ] . 中国电机工程学报 , 2019 , 39 ( 15 ): 4558 - 4568 .

XIAO Xiong , WANG Jianxiang , ZHANG Yongjun , et al . A two-dimensional convolutional neural network optimization method for bearing fault diagnosis [J ] . Proceedings of the CSEE , 2019 , 39 ( 15 ): 4558 - 4568 .

CHEN Zhiqiang , LI Chuan , SANCHEZ R V . Gearbox fault identification and classification with convolutional neural networks [J ] . Shock and Vibration , 2015 , 2015 ( 1 ): 390134 .

宫文峰 , 陈辉 , 张美玲 , 等 . 基于深度学习的电机轴承微小故障智能诊断方法 [J ] . 仪器仪表学报 , 2020 , 41 ( 1 ): 195 - 205 .

GONG Wenfeng , CHEN Hui , ZHANG Meiling , et al . Intelligent diagnosis method for incipient fault of motor bearing based on deep learning [J ] . Chinese Journal of Scientific Instrument , 2020 , 41 ( 1 ): 195 - 205 .

MONTEIRO R , BASTOS-FILHO C , CERRADA M , et al . Convolutional neural networks using Fourier transform spectrogram to classify the severity of gear tooth breakage [C ] // 2018 International Conference on Sensing, Diagnostics, Prognostics, and Control (SDPC) . Piscataway, NJ, USA : IEEE , 2018 : 490 - 496 .

王贡献 , 付泽 , 胡志辉 , 等 . 基于多分支卷积神经网络的轴承变工况故障诊断 [J ] . 噪声与振动控制 , 2023 , 43 ( 6 ): 135 - 141 .

WANG Gongxian , FU Ze , HU Zhihui , et al . A multi-branch convolution neural network for fault diagnosis of rolling bearings under variable working conditions [J ] . Noise and Vibration Control , 2023 , 43 ( 6 ): 135 - 141 .

叶壮 , 余建波 . 基于多通道一维卷积神经网络特征学习的齿轮箱故障诊断方法 [J ] . 振动与冲击 , 2020 , 39 ( 20 ): 55 - 66 .

YE Zhuang , YU Jianbo . Gearbox fault diagnosis based on feature learning of multi-channel one-dimensional convolutional neural network [J ] . Journal of Vibration and Shock , 2020 , 39 ( 20 ): 55 - 66 .

ZHANG Wei , LI Chuanhao , PENG Gaoliang , et al . A deep convolutional neural network with new training methods for bearing fault diagnosis under noisy environment and different working load [J ] . Mechanical Systems and Signal Processing , 2018 , 100 : 439 - 453 .

丁丁 , 刘文哲 , 盛常冲 , 等 . 神经网络架构搜索研究进展与展望 [J ] . 国防科技大学学报 , 2023 , 45 ( 6 ): 100 - 131 .

DING Ding , LIU Wenzhe , SHENG Changchong , et al . State of the art and prospects of neural architecture search [J ] . Journal of National University of Defense Technology , 2023 , 45 ( 6 ): 100 - 131 .

汪祖民 , 张志豪 , 秦静 , 等 . 基于卷积神经网络的机械故障诊断技术综述 [J ] . 计算机应用 , 2022 , 42 ( 4 ): 1036 - 1043 .

WANG Zumin , ZHANG Zhihao , QIN Jing , et al . Review of mechanical fault diagnosis technology based on convolutional neural network [J ] . Journal of Computer Applications , 2022 , 42 ( 4 ): 1036 - 1043 .

吴宣勇 , 黄忠全 , 李琪康 , 等 . 无源数据约束下多源域自适应的风电齿轮箱故障诊断方法 [J ] . 太阳能学报 , 2024 , 45 ( 4 ): 238 - 246 .

WU Xuanyong , HUANG Zhongquan , LI Qikang , et al . Multi-source domain adaptive fault diagnosis method of wind turbine gearbox under no-accessing source data constraints [J ] . Acta Energiae Solaris Sinica , 2024 , 45 ( 4 ): 238 - 246 .

陈建蓉 . 新型多源特征融合模型及应用研究 [D ] . 南京 : 南京大学 , 2020 .

CHUNG J , GULCEHRE C , CHO K H , et al . Empirical evaluation of gated recurrent neural networks on sequence modeling [J ] . Eprint ArXiv , 2014 . DOI: 10.48550/arXiv.1412.3555 https://doi.org/10.48550/arXiv.1412.3555 .

陈涛 , 李欣 . 太赫兹光谱在转基因菜籽油鉴别中的应用：基于改进蜉蝣算法的支持向量机模型 [J ] . 物理学报 , 2024 , 73 ( 5 ): 360 - 368 .

CHEN Tao , LI Xin . Application of terahertz spectroscopy in identification of transgenic rapeseed oils: a support vector machine model based on modified mayfly optimization algorithm [J ] . Acta Physica Sinica , 2024 , 73 ( 5 ): 360 - 368 .

GAO Zhongke , WANG Zibo , MA Chao , et al . A wavelet time-frequency representation based complex network method for characterizing brain activities underlying motor imagery signals [J ] . IEEE Access , 2018 , 6 : 65796 - 65802 .

王眺 . 基于深度交互适配网络的通用多模态学习方法研究 [D ] . 杭州电子科技大学 , 2024 .

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

一种具有强抗噪性的深度学习故障诊断模型

利用改进卷积神经网络的滚动轴承变工况故障诊断方法

采用参数化解调的变转速下柱塞泵故障诊断方法

一维卷积神经网络在往复式压缩机气阀故障诊断中的应用

采用多尺度Mamba架构的装备剩余寿命预测方法