基于振动信号的风电机组轴承故障诊断研究

发布时间：2019-01-04 09:12

【摘要】：我国风力发电近年来发展十分迅速,但高额的运行维修成本一直是影响风电场经济性的重要因素之一。轴承作为风电机组的关键部件,由于承受复杂的载荷和处于特殊的工作环境中,是导致风电机组故障停机的主要来源。因此,本文对风电机组轴承的故障诊断方法进行研究,以便能够及时准确发现轴承的早期故障,并通过采取有效措施来避免可能发生的重大事故,降低风电场的运行维修成本,增强机组的运行可靠性。本文针对风电机组的振动监测数据,提出了两种基于振动信号的风电机组轴承故障诊断算法:(1)提出一种基于变分模态分解(Variational Mode Decomposition,VMD)和Teager能量算子(Teager Energy Operator,TEO)的风电机组轴承故障诊断方法,即VMD-TEO方法。该方法首先利用变分模态分解方法对风机振动信号进行处理,分解得到若干本征模态函数分量(Intrinsic Mode Function,IMF)。然后利用峭度指标筛选出敏感IMF并计算其Teager能量算子输出,增强信号中由故障引起的冲击成分。最后对Teager能量算子输出进行傅里叶变换,获得敏感IMF的Teager能量谱,来提取故障特征频率从而进行故障诊断。(2)提出了一种基于阶次分析的风电机组故障诊断方法,该方法是对第一种方法的深化和扩展,考虑了风电机组的转速变化对轴承故障诊断的影响。该方法首先采用阶次分析技术对风电机组变转速工况下的振动信号进行角域重采样,将时域内的非平稳振动信号转换成角域内的平稳信号,再利用VMD-TEO方法分析角域平稳信号获得阶次谱,来提取故障特征阶次来实现风电机组轴承的故障诊断。在Matlab环境下,编程实现了以上算法,并对仿真信号和实验信号进行了分析,结果表明,本文提出的VMD-TEO方法和阶次分析方法均能有效准确地诊断风电机组轴承故障,并且与传统的傅里叶变换、包络谱分析以及EMD方法的结果相比,本文方法具有明显的优势。
[Abstract]:Wind power generation in China has developed rapidly in recent years, but the high cost of operation and maintenance has always been one of the important factors affecting the economy of wind farm. As a key component of wind turbine, bearing is the main source of wind turbine failure due to its complicated load and special working environment. Therefore, in this paper, the fault diagnosis method of bearing of wind turbine is studied in order to find the early fault of bearing in time and accurately, and to take effective measures to avoid possible serious accidents and reduce the operation and maintenance cost of wind farm. Enhance the operating reliability of the unit. Based on the vibration monitoring data of wind turbine, two fault diagnosis algorithms for wind turbine bearing based on vibration signal are proposed in this paper: (1) A new method based on variational mode decomposition (Variational Mode Decomposition,VMD) and Teager energy operator (Teager Energy Operator,) is proposed. TEO), the VMD-TEO method of wind turbine bearing fault diagnosis. In this method, the vibration signal of fan is processed by variational mode decomposition method, and some eigenmode function components (Intrinsic Mode Function,IMF) are obtained. Then the sensitive IMF is selected by using kurtosis index and the output of its Teager energy operator is calculated to enhance the impulse component caused by the fault in the signal. Finally, the output of Teager energy operator is transformed by Fourier transform, and the Teager energy spectrum of sensitive IMF is obtained to extract the fault characteristic frequency for fault diagnosis. (2) A fault diagnosis method for wind turbine based on order analysis is proposed. This method is an extension of the first method and takes into account the influence of the speed change of wind turbine on bearing fault diagnosis. In this method, the order analysis technique is used to resample the vibration signal of wind turbine under variable speed condition in angle domain, and the non-stationary vibration signal in time domain is converted into stationary signal in angle domain. Then the VMD-TEO method is used to analyze the angle region stationary signal to obtain the order spectrum to extract the fault characteristic order to realize the fault diagnosis of the wind turbine bearing. In Matlab environment, the algorithm is realized by programming, and the simulation signal and experimental signal are analyzed. The results show that the VMD-TEO method and the order analysis method proposed in this paper can diagnose the bearing fault of wind turbine unit effectively and accurately. Compared with the traditional Fourier transform, envelope spectrum analysis and EMD method, this method has obvious advantages.
【学位授予单位】：华北电力大学
【学位级别】：硕士
【学位授予年份】：2017
【分类号】：TM315

【参考文献】