基于小波包的EITD风力发电机组齿轮箱故障诊断  

Fault Diagnosis of Wind Turbine Gearbox Based on EITD-WPT Method

作　　者：向玲[1] 鄢小安 

机构地区：[1]华北电力大学能源动力与机械工程学院,河北保定071003

出　　处：《动力工程学报》2015年第3期205-212,共8页Journal of Chinese Society of Power Engineering

基　　金：国家自然科学基金资助项目(11072078);河北省自然科学基金资助项目(E2013502226)

摘　　要：基于三次样条插值和固有时间尺度分解中的线性变换,提出了集成固有时间尺度分解(EITD)方法,将该方法与小波包变换相结合,实现了风电机组齿轮箱故障的精确诊断.首先使用三次样条插值拟合基线控制点,将振动信号分解为一系列固有旋转分量;然后选择相关系数最大的PR分量进行小波包分解,计算分解后小波包系数的能量分布,选择能量比重较大的小波包系数重构PR分量;最后计算重构PR分量的关联维数,实现振动信号的故障诊断.利用所提出的方法对风电机组齿轮箱振动信号进行了分析,结果表明:与经验模态分解(EMD)方法处理后直接计算关联维数和经小波包的EMD方法处理后计算关联维数相比,采用小波包的EITD方法处理后计算关联维数更具有区分性,可有效识别齿轮的工作状态和故障类型.An ensemble intrinsic time-scale decomposition （EITD） method was proposed based on the cubic spline interpolation and linear transformation of intrinsic time-scale decomposition （ITD）, combining which with wavelet packet transform （WPT）, fault diagnosis of wind turbine gearbox can be realized. First, cubic spline interpolation is used to fit the baseline control points and decompose the vibration signal into several proper rotation components （PRCs）. Then, wavelet packet transform is applied to decompose the PR component of the biggest correlation coefficient and to compute the energy distribution of wavelet packet coefficients, and those wavelet packet coefficients with relatively bigger energy are selected to re- construct the PR component. Finally, correlation dimensions of the reconstructed PR component are calcu- lated and fault diagnosis of the vibration signals are achieved. The method was applied to the analysis of vi- bration signals in a wind turbine gearbox. Results show that the correlation dimension of the vibration sig- nals processed with EITD-WPT can be used to identify the working states and fault types of wind turbine gearbox more effectively, compared with that with EMD and EMD-WPT.

关 键 词：固有时间尺度分解方法 小波包分解 PR分量 关联维数 故障诊断 

分 类 号：TK83[动力工程及工程热物理—流体机械及工程]