航空发动机高低压转子间碰摩故障特征提取

针对变分模态分解（VMD）需要人为确定输入参数的问题，提出了一种参数自适应VMD（APVMD）方法，其通过粒子群（PSO）算法优化VMD的输入参数。使用APVMD结合切片谱的方法，对双转子航空发动机振动正常和异常状态下的实测弹性支承应变信号进行分析。分析结果表明，APVMD方法可以自适应地实现对实测弹性支承应变信号的分解；当发动机振动正常时，高低压转子弹性支承应变信号的APVMD模态分量中只存在高低压转子的1倍频和2倍频成分；当发动机振动异常时，低压转子弹性支承应变信号的APVMD模态分量除了包含低压转子的1倍频及倍频成分，还包含高压转子的1倍频和分频成分，以及高低压转子间的调制频率和组合频率成分；在高压转子弹性支承应变信号的APVMD模态分量中，除了高压转子的1倍频和2倍频成分，还存在高压转子的分频成分。

Feature Extraction of Rub Impact Fault Between High and Low Pressure Rotors of Aero-Engine

In order to solve the problem that the input parameters of the variational mode decomposition (VMD) need to be determined artificially, an adaptive parameter VMD (APVMD) method is proposed, in which the input parameters of VMD are optimized by particle swarm optimization (PSO) algorithm. In this paper, the measured elastic support stain signals of double rotor aero-engine under normal and abnormal vibration conditions are analyzed by using APVMD combining with slice spectrum method. The results show that the APVMD method can adaptively decompose the measured elastic support strain signals. When the aero-engine vibration is normal, the APVMD modal components of the elastic support strain signals of the high and low pressure rotors only contain 1X and 2X super-harmonic components of the high-pressure and low-pressure rotors. When the aero-engine is abnormal, the APVMD modal components of the elastic support strain signal of the low-pressure rotor include not only 1X and nX super-harmonic components of the low-pressure rotor, but also 1X super-harmonic component and sub-harmonic components of the high-pressure rotor and the modulation frequency and combined frequency components between the high-pressure rotor and low-pressure rotor. In the APVMD modal component of the high pressure rotor elastic support strain signal, in addition to the 1X and 2X super-harmonic components of the high-pressure rotor, there are also sub-harmonic components of the high-pressure rotor.