摘要
针对某泵站立式混流泵机组异常振动现象,采用低频振动传感器和电涡流传感器,分别测试开机过程和正常运行时水泵机组下机架、泵盖、外筒体振动以及主轴摆度.基于希尔伯特-黄变换进行试验数据处理,获得了测试信号位移峰峰值,通过提取稳定运行时的振动信号特征向量,对故障类型进行识别;在此基础上,通过分析开机过程信号时频变化情况,寻找诱发机组异常振动的故障原因.现场振动测试分析结果表明:稳定运行时机组振动频率主要集中在转频附近,且主轴轴心轨迹成椭圆形,可诊断其故障类型属于转子不平衡;在开机过程中,主轴的轴心轨迹紊乱,开机瞬间机组的振动值为正常运行时振动值的47.3倍,据此可判断出在此刻机组内部产生了强大的冲击力,进而引起转子不平衡运转.采用希尔伯特-黄变换法能够准确获取水泵振动信号的时频特征,尤其对分析处理高度非线性的被测信号具有很强的优势,而开展开机过程中的振动测试研究,也为水泵故障诊断分析方法开辟了一条新途径.
To identify the reasons for the abnormal vibration in a vertical mixed-flow pump,the vertical and horizontal vibrations were measured at the low bracket,pump lid and outside column of pump unit by using low-frequency vibration sensors in startup and normal operation periods,the pump shaft run out were monitored by means of an eddy current sensor as well.The experimental data were processed by the Hilbert-Huang transform（HHT）,and the vibration peak-peak displacements were acquired,and the eigenvectors of vibration signals were extracted to diagnose the type of fault;finally,the time-frequency feature in startup period was analyzed in order to search the root reasons for the abnormal vibration.The analyzed results showed that the vibration frequency of the pump unit was close to the unit rotational frequency in the normal operation;additionally,for the shaft orbit is similar to an eccentric ellipse,the type of fault in the unit was unlikely the rotor imbalance.During the starting period,the shaft orbit was chaos and the vibration amplitude was as large as 47.3 times the amplitude in the normal operation,suggesting a violent force might be generated in the starting period and eventually induces the rotor to operate in imbalance.The time-frequency characteristics of vibration signals in a pump can be acquired accurately with the HHT,especially for highly nonlinear signals;meanwhile,this study on vibration gene-rated in a pump during its startup period will provide a new way for pump troubleshooting.
出处
《排灌机械工程学报》
EI
北大核心
2013年第5期373-378,共6页
Journal of Drainage and Irrigation Machinery Engineering
基金
国家自然科学基金资助项目(50879062)
教育部博士点基金资助项目(20090141110057)
关键词
混流泵
振动
故障诊断
希尔伯特-黄变换法
时频分析
mixed-flow pump
vibration
fault diagnosis
Hilbert-Huang transform
time-frequency analysis
