基于VMD倒频变换的风电机组齿轮箱复合故障诊断  

Composite fault diagnosis of wind turbine gearboxes based on VMD cepstral transform

作　　者：吴岚 柳亦兵[1] 吴仕明 任锦胜 滕伟[1] WU Lan;LIU Yibing;WU Shiming;REN Jinsheng;TENG Wei(Key Laboratory of Power Station Energy Transfer Conversion System,Ministry of Education,North China Electric Power University,Beijing 102206,China;Beijing Envada Power Electronics Engineering Technology Ltd.,Beijing 100086,China)

机构地区：[1]华北电力大学电站能量传递转化与系统教育部重点实验室,北京102206 [2]北京英华达电力电子工程科技有限公司,北京100086

出　　处：《振动与冲击》2023年第24期221-227,256,共8页Journal of Vibration and Shock

摘　　要：风电机组齿轮箱结构复杂,零部件多,可能产生复合故障。同一旋转轴上的不同零部件(齿轮、轴承等)的故障信息往往具有相同的转频边带频率成分,分布在信号频谱的不同频带中,倒频谱对边带频率信息进行了压缩处理,不能识别此类复合故障;基于窄带倒谱变换的思想,将变分模态分解(variational mode decomposition, VMD)应用到振动信号的对数谱中,提出了一种基于变分模态分解VMD倒频变换的方法,实现了对这一复合故障的精细诊断。通过仿真信号和风电场实例信号的验证,相比于窄带倒谱变换,该方法在有效区分此类风电机组齿轮箱复合故障的基础上,改善了频域和倒频域的分辨率,拥有更高的诊断定位精度。Wind turbine gearboxes have complex structures and many parts,which may generate compound faults.The fault information of different parts(gears,bearings,etc.)on the same rotating shaft often has the same sideband components,distributed in different frequency bands of the signal frequency spectrum.The cepstrum compresses the sideband information and cannot identify such compound faults.Based on the idea of narrow-band cepstrum transformation,the variational modal decomposition(VMD)was applied to the logarithmic spectrum to achieve a fine diagnosis of this composite fault in this paper.Through the verification of simulated signals and actual signals from wind farms,the method is effective in distinguishing such wind turbine gearbox compound faults.Compared with the narrow-band cepstrum transformation,the method improves the resolutions of frequency domain and cepstrum domain,and possesses higher diagnostic localization accuracy.

关 键 词：变分模态分解(VMD) 倒频谱 复合故障 风电机组 

分 类 号：TH165.3[机械工程—机械制造及自动化]