摘要
风电齿轮箱是风机传动链的重要组成部分,其中高速级齿轮工作环境复杂多变,失效破坏易发,极易引发齿轮箱的故障,甚至带来灾难性后果而停机。实时状态监测及故障诊断,对于确保风机正常运行尤为重要。基于振动信号的风机故障诊断研究方法层出不穷,然而现有的方法对具体型号风机不具有普适性。因此,该文对在齿轮失效特征归类总结的基础上,选取高速齿轮断齿时的振动信号,通过小波变换对原信号消噪,有效辨识边频带特征,以反映齿轮失效程度;进而通过频谱变换获得频域特征,以反映齿轮失效的类型。结果表明,断齿发生时,时域上幅值有周期性冲击,周期为转速的倒数。频域上,转频的5倍频均有大幅增加,甚至达到127倍;高转速下啮合频率2倍边频带不对称性更为明显,3倍频幅值增加达4. 4倍;断齿失效下会引起该型号风机1 200~1 500 Hz频段的共振现象,研究结果可作为该型号风机断齿失效诊断的基础数据。
Wind power gearbox is an important part of fan transmission chain,in which the high-speed gear working environment is complex and variable.The failure and damage of wind power gearbox is easy to cause gearbox fault,or even bring catastrophic consequences and stop.Real-time status monitoring and fault diagnosis are particularly important to ensure normal operation of the fan.There were many research methods for fan fault diagnosis based on vibration signals,but the existing methods were not universal for specific types of fans.Therefore,based on classification and summary of gear failure characteristics,vibration signal gear was selected to analysis broken failure in high speed.Original signal was de-noised by wavelet transform,and edge frequency band characteristics were effectively identified to reflect the degree of gear failure.Then frequency domain characteristics were obtained by frequency spectrum transformation to reflect type of gear failure.Results showed that the upper amplitude of time domain has periodic impact and the period was reciprocal of rotational speed.In frequency domain,frequency of 5 times of frequency of revolution has been greatly increased,even reaching 127 times.At high rotation speed,asymmetry of mesh frequency band of 2 times side frequency was more obvious,and amplitude of 3 times frequency increases by 4.4 times.The failure of broken teeth will cause resonance phenomenon in frequency range of 1 200 Hz to 1 500 Hz of this type of fan.
作者
慕松
许辉
王春秀
宿友亮
MU Song;XU Hui;WANG Chunxiu;SU Youliang(Institute of Mechanical Engineering,Ningxia University,Yinchuan Ningxia 750021,China)
出处
《农业工程》
2019年第2期75-80,共6页
AGRICULTURAL ENGINEERING
基金
基于数据驱动的风电机组关键部件故障预测及健康管理系统开发与示范应用(项目编号:2018YBZD1644)
关键词
轮齿折断
小波消噪
特征提取
gear broken
wavelet de-noising
feature extraction