Fault diagnosis of rolling bearing of wind turbine generator based on PSO-SEBD
WANG Pengcheng1,2, DENG Aidong1,2, LING Feng1,2, DENG Minqiang1,2, LIU Yang1,2
1. National Engineering Research Center Power Generation Control and Safety, Southeast University, Nanjing 210096, China;
2. School of Energy and Environment, Southeast University, Nanjing 210096, China
摘要滚动轴承作为风电机组传动系统的关键部件，其健康状态监测对整个机组的安全稳定运行至关重要。针对滚动轴承的故障诊断问题，在基于先验未知盲反卷积技术的包络谱重复瞬态循环平稳性提取方法(extracting cyclo-stationarity of repetitive transients from envelope spectrum based on prior-unknown blind deconvolution technique, SEBD)的基础上，提出了一种基于粒子群算法(particle swarm optimization, PSO)寻优的SEBD滚动轴承故障诊断方法，实现SEBD滤波器长度自适应选择。首先，以最大故障特征频率比(characteristic frequency ratio, CFR)作为适应度函数，利用PSO算法对滤波器长度进行寻优；然后，利用获得的最优滤波器长度进行SEBD处理；最后，根据SEBD处理后信号的包络谱特征实现轴承故障的有效识别。通过对仿真信号和德国帕德博恩大学公开轴承故障数据进行分析，验证了PSO-SEBD的有效性。通过与几种常用的诊断方法对比以及噪声环境下分析，表明该方法具有较好的诊断性能和抗噪声能力。
Abstract：As a key component of the transmission system of a wind turbine, the rolling bearing is vital to the safe and stable operation of the entire unit. Aiming at the problem of rolling bearing fault diagnosis, On base of extracting cyclo-stationarity of repetitive transients from envelope spectrum based on prior-unknown blind deconvolution technique, a SEBD bearing fault diagnosis method based on particle swarm optimization (PSO) optimization is proposed to realize the adaptive selection of SEBD filter length. Firstly, take the maximum fault characteristic frequency ratio (CFR) as the fitness function, and use the PSO algorithm to optimize the filter length; then, use the obtained optimal filter length to perform SEBD processing; finally, according to the envelope spectrum feature of the signal after SEBD processing realizes the effective identification of bearing faults. The effectiveness of PSO-SEBD is verified by analyzing the simulation signal and the public bearing failure data of Paderborn University in Germany. By comparing with several commonly used diagnostic methods and analyzing under noisy environment, it shows that this method has better diagnostic performance and anti-noise ability.
王鹏程1,2，邓艾东1,2，凌峰1,2，邓敏强1,2，刘洋1,2. 基于PSO-SEBD的风电机组滚动轴承故障诊断[J]. 振动与冲击, 2023, 42(7): 281-288.
WANG Pengcheng1,2, DENG Aidong1,2, LING Feng1,2, DENG Minqiang1,2, LIU Yang1,2. Fault diagnosis of rolling bearing of wind turbine generator based on PSO-SEBD. JOURNAL OF VIBRATION AND SHOCK, 2023, 42(7): 281-288.
[1] 向玲,王朋鹤,李京蓄.基于CNN-LSTM的风电机组异常状态检测[J].振动与冲击,2021,40(22):11-17.
XIANG Ling, WANG Peng-he, LI Jing-xu. Detection of abnormal state of wind turbine based on CNN-LSTM [J]. Vibration and Shock, 2021, 40(22): 11-17.
[2] 齐咏生,樊佶,李永亭等.基于增强型形态学滤波的风电机组轴承故障诊断方法[J].振动与冲击,2021,40(04):212-220.
QI Yong-sheng, FAN Ji, LI Yong-ting, etc. A fault diagnosis method for wind turbine bearings based on enhanced morphological filtering [J]. Vibration and Shock, 2021, 40(04): 212-220.
[3] Bp A , Mn B , Gt C . Recursive variational mode extraction and its application in rolling bearing fault diagnosis[J]. Mechanical Systems and Signal Processing, 2022, 165:108321-.
[4] Qin, Yi. A New Family of Model-Based Impulsive Wavelets and Their Sparse Representation for Rolling Bearing Fault Diagnosis[J]. IEEE TRANSACTIONS ON INDUSTRIAL ELECTRONICS, 2017:1-1.
[5] 刘志川,唐力伟,曹立军.基于MED及FSK的滚动轴承微弱故障特征提取[J].振动与冲击,2014,33(14):137-142.
LIU Zhi-chuan, TANG Li-wei, CAO Li-jun. Feature extraction of weak faults of rolling bearings based on MED and FSK [J]. Vibration and Shock, 2014, 33(14): 137-142.
[6] 胡爱军,严家祥,白泽瑞.基于MOMEDA和增强倒频谱的风电机组齿轮箱多故障诊断方法[J].振动与冲击,2021,40(07):268-273.
HU Ai-jun, YAN Jia-xiang, BAI Ze-rui. Multi-fault diagnosis method of wind turbine gearbox based on MOMEDA and enhanced cepstrum [J]. Vibration and Shock, 2021, 40(07): 268-273.
[7] 吕中亮,汤宝平,周忆等.基于网格搜索法优化最大相关峭度反卷积的滚动轴承早期故障诊断方法[J].振动与冲击,2016,35(15):29-34.
LV Zhong-liang, TANG Bao-ping, ZHOU Yi, etc. Based on the grid search method to optimize the maximum correlation kurtosis deconvolution method for early fault diagnosis of rolling bearings [J]. Vibration and Shock, 2016, 35(15): 29-34.
[8] 王新刚,王超,韩凯忠. 基于优化VMD和MCKD的滚动轴承早期故障诊断方法[J]. 东北大学学报(自然科学版), 2021, 42(3): 373-380, 388.
WANG Xin-gang, WANG Chao, HAN Kai-zhong. Early fault diagnosis method of rolling bearing based on optimized VMD and MCKD[J]. Journal of Northeastern University (Natural Science Edition), 2021, 42(3): 373-380, 388.
[9] 齐咏生,刘飞,李永亭,等. 基于MK-MOMEDA和Teager能量算子的风电机组滚动轴承复合故障诊断[J]. 太阳能学报, 2021, 42(7): 297-307.
QI Yong-sheng, LIU Fei, LI Yong-ting, et al. Composite Fault Diagnosis of Wind Turbine Rolling Bearings Based on MK-MOMEDA and Teager Energy Operator [J]. Chinese Journal of Solar Energy, 2021, 42(7): 297-307.
[10] 张俊,张建群,钟敏,等. 基于PSO-VMD-MCKD方法的风机轴承微弱故障诊断[J]. 振动.测试与诊断, 2020, 40(2): 287-296, 418.
ZHANG Jun, ZHANG Jian-qun, Zhong Min, et al. Weak fault diagnosis of fan bearings based on PSO-VMD-MCKD method [J]. Vibration. Testing and Diagnosis, 2020, 40(2): 287-296, 418.
[11] He L , Wang D , Yi C , et al. Extracting cyclo-stationarity of repetitive transients from envelope spectrum based on prior-unknown blind deconvolution technique[J]. Signal Processing, 2021, 183:107997.
[12] Deng M , Zhu J , Shi Y , et al. Resonance-based bandwidth Fourier decomposition method for gearbox fault diagnosis[J]. Measurement Science and Technology, 2021, 32(3):035003 (15pp).
[13] 陈昆弘, 刘小峰. 基于循环相关和LPSO算法的自适应MCKD方法的滚动轴承早期故障特征提取[J]. 2022(22).
CHEN Kun-hong, LIU Xiao-feng. Adaptive MCKD method based on cyclic correlation and LPSO algorithm for early fault feature extraction of rolling bearings [J]. 2022(22).
[14] Jia X , Ming Z , Yuan D , et al. Sparse filtering with the generalized Lp/Lq norm and its applications to the condition monitoring of rotating machinery[J]. Mechanical Systems & Signal Processing, 2017, 102C(MAR.1):198-213.
[15] Feng Z , Ma H , Zuo M J . Vibration signal models for fault diagnosis of planet bearings[J]. Journal of Sound & Vibration, 2016, 370:372-393.
[16] Deng M , Deng A , Zhu J , et al. Adaptive bandwidth Fourier decomposition method for multi-component signal processing[J]. IEEE Access, 2019, PP(99):1-1.
