基于ALIF-HT的汽轮发电机组转子故障诊断

发布时间：2018-03-14 12:55

本文选题：汽轮发电机组　切入点：转子　出处：《动力工程学报》2017年11期 　论文类型：期刊论文

【摘要】：针对汽轮发电机组转子故障振动信号为多分量非平稳信号,将一种新的信号分解方法——自适应局部迭代滤波(ALIF)用于转子故障振动信号分解,并与希尔伯特变换(HT)相结合,提出了基于ALIF-HT的汽轮发电机组转子故障诊断方法:首先对转子原始振动信号进行ALIF得到若干信号分量,再应用HT求取每个分量的瞬时频率,获取原信号全部信号分量的完整时频表示,最后根据转子故障振动信号的时频特征判别转子的故障类型.通过仿真信号分析验证ALIF对多分量信号的分解能力,并利用转子油膜失稳故障分析验证该方法的工程实用性.结果表明:ALIF方法能够有效克服经验模态分解(EMD)存在的模态混叠问题,使得ALIF-HT方法相对于希尔伯特黄变换(HHT)方法具有更高的时频分析精度.
[Abstract]:A new signal decomposition method, adaptive local iterative filter (ALIFF), is used to decompose the rotor fault vibration signal, which is based on the multi-component non-stationary signal of turbine generator rotor fault, and is combined with Hilbert transform (HTT). A fault diagnosis method of turbine generator rotor based on ALIF-HT is proposed. Firstly, some signal components are obtained by ALIF for the original vibration signal of the rotor, and then the instantaneous frequency of each component is obtained by using HT. The complete time-frequency representation of all the signal components of the original signal is obtained. Finally, the fault types of the rotor are identified according to the time-frequency characteristics of the rotor fault vibration signal. The decomposition ability of the ALIF to the multi-component signal is verified by the simulation signal analysis. The engineering practicability of this method is verified by the fault analysis of rotor oil film instability. The results show that the proposed method can effectively overcome the modal aliasing problem existing in empirical mode decomposition (EMD). The ALIF-HT method has higher time frequency analysis accuracy than the Hilbert-Huang transform (HHT) method.
【作者单位】：华北电力大学能源动力与机械工程学院;
【基金】：国家自然科学基金资助项目(51307058) 河北省自然科学基金资助项目(E2014502052) 中央高校基本科研业务费专项基金资助项目(2017XS134)
【分类号】：TM311

【相似文献】