基于RSAN和EEMD的MKSVM多模态融合旋转机械故障分类

刘东1, 2 谭鋆1, 2 龙小波1, 2 赵杰3, 赵志高4

振动与冲击 ›› 2025, Vol. 44 ›› Issue (9) : 290-302.

PDF(5228 KB)
PDF(5228 KB)
振动与冲击 ›› 2025, Vol. 44 ›› Issue (9) : 290-302.
故障诊断分析

基于RSAN和EEMD的MKSVM多模态融合旋转机械故障分类

  • 刘东*1, 2 谭鋆1, 2 龙小波1, 2 赵杰3, 赵志高4
作者信息 +

MKSVM multimode fusion rotating machinery fault classification method based on RSAN and EEMD

  • LIU Dong*1,2, TAN Yun1,2, LONG Xiaobo1,2, ZHAO Jie3, ZHAO Zhigao4
Author information +
文章历史 +

摘要

针对真实场景下旋转机械的故障振动信号存在样本数量不足和具备较强复杂性的问题,思考单一模态下故障诊断的限制,提出一种基于残差空间注意力网络(residual spatial attention network,RSAN)和集合经验模式分解(ensemble empirical mode decomposition,EEMD)的多核支持向量机(multi-kernel support vector machine,MKSVM)多模态融合旋转机械故障分类方法。首先,对原始数据进行小波阈值去噪处理,将降噪的一维信号转化为二维递归图;然后,设计RSAN结构,输入训练集的递归图进行训练,捕获故障信号的递归特征;随后,计算EEMD的固有模态函数(intrinsic mode function,IMF)的样本熵,基于其变异系数来选取故障信号的时域特征。最后,利用MKSVM对多模态特征向量进行高维映射和融合,完成旋转机械的故障分类。结果表明,在少样本的转子和轴承故障数据上准确率分别达到100%和98.3%,证明该方法具备良好的可行性。

Abstract

In order to solve the problems of insufficient samples and high complexity of faulty vibration signals of rotating machines in real scenarios, a multimodal fusion rotating machine fault classification method with multi-kernel support vector machine (MKSVM) based on residual space attention network (RSAN) and ensemble empirical mode decomposition (EEMD) was proposed. First, wavelet thresholding denoising was performed on the raw data to transform the noise-reduced one-dimensional signal into a two-dimensional recurrence map; Then, the RSAN structure was designed to be trained by inputting the recurrence graph of the training set to capture the recurrence features of the fault signals; Subsequently, the sample entropy of the intrinsic modal function (IMF) of the EEMD was calculated and the time-domain characteristics of the fault signal were selected based on its coefficient of variation. Finally, a MKSVM was used to perform high-dimensional mapping and fusion of multimodal feature vectors to complete the fault classification of rotating machinery. It was shown that the accuracy reaches 100% and 98.3% prediction performance on rotor and bearing failure data with few samples, respectively, which proves that the method possesses good feasibility.

关键词

故障诊断 / 残差网络 / 注意力机制 / 支持向量机 / 特征融合

Key words

fault diagnosis; residual network;  / attention mechanism;  / support vector machine; feature fusion.

引用本文

导出引用
刘东1, 2 谭鋆1, 2 龙小波1, 2 赵杰3, 赵志高4. 基于RSAN和EEMD的MKSVM多模态融合旋转机械故障分类[J]. 振动与冲击, 2025, 44(9): 290-302
LIU Dong1, 2, TAN Yun1, 2, LONG Xiaobo1, 2, ZHAO Jie3, ZHAO Zhigao4. MKSVM multimode fusion rotating machinery fault classification method based on RSAN and EEMD[J]. Journal of Vibration and Shock, 2025, 44(9): 290-302

参考文献

[1]  Sun Jiedi,Yan Changhong,Wen Jiangtao. Intelligent bearing fault diagnosis method combining compressed data acquisition and deep Learning[J]. IEEE Transactions on Instrumentation and Measurement,2018,67(1):185–195.
[2]  Cheng W,Zhang Z,Zhang J,Lu J. Acoustical source tracing using independent component analysis and correlation analysis[J]. Shock and Vibration,2015,2015(1):571206.
[3]  S.P. Mogal,D.I. Lalwani,Experimental investigation of unbalance and misalignment in rotor bearing system using order analysis,Journal of Measurements in Engineering,2015,3(4):114–122.
[4] 刘运航,宋宇博,朱大鹏.中心修正投影结合IGWO-SVM的滚动轴承故障分类方法[J].振动与冲击,2023,42(24):267-275.
LIU Yunhang, SONG Yubo, ZHU Dapeng. A rolling bearing fault classification method based on IGWO-SVM combined with center correction projection[J]. Journal of Vibration and Shock, 2023, 42(24): 267-275.
[5]  D. Liu,H.T. Zeng,Z.H. Xiao,L.H. Peng,O.P. Malik,Fault diagnosis of rotor using EMD thresholding-based de-noising combined with probabilistic neural network[J],Journal of Vibroengineering,2017,19(8):5920–5931.
[6]  B.K. Alpert,A class of bases in L2 for the sparse representation of integraloperators,SIAM Journal on Mathematical Analysis, 1993, 24(1):246-262..
[7] 张俊红,李林洁,马文朋,等.EMD-ICA联合降噪在滚动轴承故障诊断中的应用[J].中国机械工程,2013,24(11):1468-1472.
Zhang Junhong, Li Linjie, Ma Wenpeng, et al. LI Zhou-Yu, LIU Yu. Application of EMD-ICA to Fault Diagnosis of Rolling Bearings[J]. China Mechanical Engineering, 2013, 24(11): 1468-1472.
[8] Jiang H, Li C, Li H. An improved EEMD with multiwavelet packet for rotating machinery multi-fault diagnosis. Mechanical Systems and Signal Processing. 2013 Apr 1;36(2):225-39.
[9]  H. Li,L. Yang,D. Huang,The study of the intermittency test filtering character of Hilbert–Huang transform.Mathematics & Computers in Simulation, 2005, 70(1):22-32..
[10] Z. Wu,N.E. Huang,Ensemble empirical mode decomposition: a noise assisted data analysis method,Advances in Adaptive Data Analysis, 2011, 1(01)1–41.
[11] 何青,褚东亮,毛新华.基于EEMD和MFFOA-SVM滚动轴承故障诊断[J].中国机械工程,2016,27(09):1191-1197.
He Qing, Chu Dongliang, Mao Xinhua. Study on Rolling Bearing Fault Diagnosis Based on EEMD and MFFOA-SVM[J]. China Mechanical Engineering, 2016,27(09):1191-1197.
[12] Y.G. Lei,Z.J. He,Y.Y. Zi,Application of the EEMD method to rotor fault diagnosis of rotating machinery, Mechanical Systems and Signal Processing, 2009, 23(4): 1327–1338.
[13] 童靳于,罗金,郑近德.基于增强深度自编码网络的滚动轴承故障诊断方法[J]. 中国机械工程, 2021, 32(21): 2617-2624.
TONG Jinyu, LUO Jin, ZHENG Jinde. Rolling Bearing Fault Diagnosis Method Based on Enhanced Deep Auto-encoder Network[J]. China Mechanical Engineering, 2021, 32(21): 2617-2624.
[14] 张继冬,邹益胜,邓佳林,等. 基于全卷积层神经网络的轴承剩余寿命预测 [J]. 中国机械工程, 2019, 30 (18): 2231-2235.
ZHANG Jidong, ZOU Yisheng, DENG Jialin, ZHANG Xiaolu, et al. Bearing Remaining Life Prediction Based on Full Convolutional Layer Neural Networks[J]. China Mechanical Engineering, 2019, 30 (18): 2231-2235.
[15] 张瑞成,白晓泽,董砚,等.基于 SSAE-LSTM神经网络的风电变流器开路故障诊断[J]. 可再生能源,2023,(41)03:361-369.
ZHANG Ruicheng, BAI Xiaoce, DONG Yan, et al. Open-Circuit Fault Diagnosis of Wind Turbine Converter Based on SSAE-LSTM Neural Network[J]. Renewable Energy Resources, 2023, (41)03: 361-369.
[16] 冯国红, 王宏恩, 刁鹏飞, 等.小样本下基于递归图和迁移学习的轴承故障诊断[J/OL]. 机床与液压,: 1-17[2024-07-17].
FENG Guoqiang, WANG Hongen, DIAO Pengfei, et al. Bearingfault diagnosis based on recursive graph and transfer learning in small sample[J]. Machine Tool & Hydraulics: 1-17[2024-07-17].
[17] 伍济刚, 文港, 杨康. 改进一维卷积神经网络的航空发动机故障诊断方法[J]. 电子测量与仪器学报, 2023, 37(3): 179-186.
WU Jigang, WEN Gang, YANG Kang. Improved One-Dimensional Convolutional Neural Network for Aero-Engine Fault Diagnosis[J]. Journal of Electronic Measurement and Instrumentation, 2023, 37(3): 179-186.
[18] 李世晓, 杜锦华, 龙云. 基于一维卷积神经网络的机电作动器故障诊断[J]. 电工技术学报, 2022, 37(S1): 62-73.
LI Shixiao, DU Jinhua, LONG Yun. One-Dimensional Convolutional Neural Network-Based Fault Diagnosis of ElectromEchanical Actuators[J]. Electrical Engineering Magazine, 2022, 37(S1): 62-73.
[19] 崔桂艳,钟倩文,郑树彬,等. 基于VMD灰度图像编码和CNN的多传感融合轴承故障诊断[J]. 振动与冲击, 2023, 42(21): 316-326.
CUI Guiyan, ZHONG Qianwen, ZHENG Shubin, et al. Multi-sensor fusion bearing fault diagnosis based on VMD gray level image coding and CNN[J]. Journal of Vibration and Shock, 2023, 42(21): 316-326.
[20] 尹爱军,吕明阳,杨敏英,等. RepVGG与CapsNet融合的轴承故障诊断[J]. 振动与冲击, 2024, 43(14): 301-307.
Aijun, L Mingyang, YANG Minying, et al. Bearing fault diagnosis based on the combined use of RepVGG and CapsNet[J]. Journal of Vibration and Shock, 2024, 43(14): 301-307
[21] 秦国浩,张楷,丁昆,等.动态宽卷积残差网络的轴承故障诊断方法[J].中国机械工程,2023,34(18):2212-2221.
QIN Guohao, ZHANG Kai, DING Kun, et al. Dynamic Wide Convolutional Residual Network for Bearing Fault Diagnosis Method[J]. China Mechanical Engineering, 2023, 34(18): 2212-2221.
[22] 邹筱瑜,孙国庆,王忠宾,等.基于时频融合深度网络的矿用钻机轴承故障诊断[J/OL].中国机械工程:1-10[2024-08-06].
ZOU Xiaoyu, SUN Guoqing, WANG Zhongbin, et al. Bearing fault diagnosis of mining drilling rig with time-frequency-fused deep network[J]. China Mechanical Engineering: 1-10[2024-08-06].
[23] 周翔宇,毛善君,李梅.基于频域降采样和CNN的轴承故障诊断方法[J].北京大学学报(自然科学版),2023,59(02):251-260.
ZHOU Xiangyu, MAO Shanjun, LI Mei. Bearing Fault Diagnosis Method Based on Down-Sampling in Frequency Domain and CNN[J]. Acta Scientiarum Naturalium Universitatis Pekinensis, 2023, 59(2): 251-260.
[24] 陈尚年,李录平,张世海,等. 基于EEMD-LSTM的汽轮机转子碰磨故障诊断模型及其工程应用[J].热能动力工程, 2023, 38(08): 159-168.
CHEN Shangnian, LI Luping, ZHANG Shihai, et al. EEMD-LSTM-based Turbine Rotor Rub-impact Fault Diagnosis Model and Its Engineering Application[J]. Journal of Engineering for Thermal Energy and Power, 2023,38(08):159-168.
[25] 蔡艳平,范宇,陈万,等. 改进时频分析和特征融合在内燃机故障诊断中的应用[J].中国机械工程,2020,31(16):1901-1911.
CAI Yanping, FAN Y, CHEN Wan, et al. Applications of Improved Time-frequency Analysis and Feature Fusion in Fault Diagnosis of IC Engines[J]. China Mechanical Engineering, 2020, 31(16): 1901-1911.
[26] Liu,Dong,et al. Feature extraction of rotor fault based on EEMD and curve code[J]. Measurement,2019,135:712-724.
[27] CWRU. The case western reserve university bearingdata center [EB/OL]. (2021–08–10) [2022–03–10]. https://engineering.case.edu/bearingdatacenter/download-data-file.
[28] 胡晓,肖志怀,刘东,等.  基于EEMD-SDCCⅠ-HMM的水电机组振动故障识别方法[J]. 振动与冲击, 2022, 41(3): 165-175.
HU Xiao, XIAO Zhihuai, LIU Dong, et al. Vibration fault identification method of hydropower unit based on EEMD-SDCCⅠ-HMM[J]. Journal of Vibration and Shock, 2022, 41(3): 165-175.
[29] 汪峰,周凤星,严保康.基于特征量融合和支持向量机的滚动轴承故障诊断[J]. 科学技术与工程, 2022, 22(6): 2351-2356.
Wang Feng, Zhou Fengxing, Yan Baokang. Rolling bearing fault diagnosis based on feature fusion and support vector machine[J]. Science Technology and Engineering, 2022,22(6):2351-2356.

PDF(5228 KB)

113

Accesses

0

Citation

Detail

段落导航
相关文章

/