转子轴承系统的振动信号常呈现非线性调频特征且信号分量在频域混叠,传统的频谱分析方法难以处理该类信号.基于参数化解调的非线性调频信号分解方法来分析油膜涡动、油膜振荡特征信号能够有效分解频域混叠的非平稳信号.首先通过优化频谱集中性指标来估计信号瞬时频率参数并用估计到的参数将非线性调频信号解调为平稳信号,最后用带通滤波器提取解调信号.仿真及实验信号通过该方法分析后的结果证明,所用非线性调频分量分解的信号分解方法能够有效提取转子轴承系统的油膜涡动、油膜振荡故障特征,从信号时频图及提取分量的时域图可以清晰看到油膜涡动、油膜振荡的发生发展过程,为早期油膜涡动判定提供依据.%The vibration signal of rotor bearing systems often shows the characteristics of nonlinear frequency modulation in non-stationary conditions. Traditional spectral analysis methods are sometimes difficult to deal with these kinds of signals which components are mixed in the frequency domain. In this paper, based on the parameter resolution, a method of the nonlinear frequency modulation signal decomposition is applied to the analysis of oil whirl and oil whip characteristics. Firstly, the instantaneous frequency parameters of the signal are estimated by optimizing the spectral concentration index. Then, with the estimated parameters, the nonlinear FM signal is used as the stationary signal. Finally, the demodulation signal is extracted with the band pass filter. Results of simulation and test of signals show that this method can effectively decompose the non-stationary signals in the frequency domain. The fault characteristics of the oil whirl and oil whip of the bearing system can be extracted effectively. The initiation and development of the oil whirl and oil whip can be observed clearly and the time-frequency and amplitude information can be accurately detected. The half-frequency oil whirl can be found clearly from time-frequency spectrum diagram and the time domain waveforms of the extracted components, which provides time information for fault diagnosis. The results provide a basis for early judgment of oil whirls of bearing systems of rotors.
展开▼
