基于改进阶次分析与自适应VMD的变转速齿轮箱故障诊断研究

Study on the Fault Diagnosis of Variable Speed Gearbox based on Improved Order Analysis and Adaptive VMD

变转速齿轮箱由于工况复杂导致转频不稳定,齿轮箱的微弱故障信号可能会被掩盖在强噪声中,不能直接应用传统的时频分析方法,为故障特征的提取增加一定的难度。针对变转速信号的处理,传统的计算阶次分析方式(COT)很好地解决了变转速齿轮箱的故障特征难以提取出来的问题,但由于传统COT中所使用的重采样方法是基于样条插值法的,无法根据转频选取转频,导致重采样间隔并不均匀;提出了改进的阶次分析方法,根据采样的各点角速度依次进行重采样,提高了阶次分析的精度。同时,变转速齿轮箱因动力传递复杂,导致变转速齿轮箱噪声更加严重。变分模态分解(VMD)常被被用来去除复杂信号噪声,提取被掩盖在强噪声中的微弱故障信号。提出了自适应VMD使用能量法,确定分解层数后对分量进行指标化选取,使去噪的效果得到进一步提升。通过对实验信号分析,使用此方法进行验证。结果表明,此方法能有效转换变转速齿轮箱信号为阶次信号并对故障进行识别。

Due to the complex operating conditions of the gearbox with variable speed,the frequency of the gearbox is unstable,the weak fault signal of the gear box may be hidden in the strong noise,it is not possible to directly apply the traditional time-frequency analysis method,which increases the difficulty of fault feature extraction.For the processing of the variable speed signal,the traditional calculation order analysis(COT)can solve the problem that the fault features of the variable speed gearbox are difficult to be extracted.However,because the resampling method used in the traditional COT is based on spline interpolation method,the frequency cannot be selected according to the frequency,resulting in uneven resampling interval.An improved order analysis is proposed,which resamples according to the angular velocity of each sampling point in order to improve the accuracy of order analysis,the accuracy of the order analysis is improved.At the same time,the noise of variable speed gearbox is more serious because of the complexity of power transmission.The variational mode decomposition(VMD)is often used to eliminate complex signal noise and extract weak fault signals hidden in strong noise.An adaptive VMD using energy method is proposed to determine the number of decomposition layers and then select the components for indexing,so that the effect of denoising is further improved.Through the analysis of the experimental signals,the method is used to verify the results.The results show that the method can effectively transform the signal of the gearbox with variable speed into the signal of order and identify the fault.