【摘要】 传统的短时傅里叶变换(Short-time Fourier transform,STFT)峰值搜索法对高噪声、强干扰信号进行瞬时频率估计的结果往往偏离真实值误差较大,且对复杂的旋转机械设备产生的振动信号不能实现参考轴的瞬时频率估计。针对旋转机械升降速阶段振动信号的特点,提出一种新的旋转机械升降速阶段振动信号瞬时频率估计算法——STFT Viterbi拟合法(STFT_Viterbi algorithm fit,STFT_VF)。该方法采用STFT对振动信号进行时频分析,从而得到时间离散点和频率离散点组成的网格面,然后运用Viterbi算法实现对参考轴信号的瞬时频率估计。STFT_VF方法极大地降低了噪声和干扰对瞬时频率估计结果的影响,实现了对复杂旋转机械振动信号的瞬时频率估计,且结果精度高。仿真和实际测试试验验证了本方法的正确性。更多还原

【Abstract】 The vibration signal of rotational machine in the station of run up and run down in the high noise environment,the traditional method of peak detection,by using short-time Fourier transform(STFT),gives inaccurate estimation.A new STFT_Viterbi algorithm fit(STFT_VF) method is proposed that can estimate such instantaneous frequency of vibration signal of rotating machinery.Estimation of the instantaneous frequency,by using the STFT,is considered.And algorithm realization is performed recursively by using the Viterbi algorithm.However,the results through direct searching are often worse in precision,due to the interference of strong noise and adjacent components of signal.The STFT_VF method is used to solve the problem.The correctness of the method based on STFT_VF is validated through simulation and test.更多还原