为解决发动机高速轴承振动数据时域信号中隐藏的突变信号难以精确时频定位的问题，提出一种突变信号的时幅拐点分析算法． 该算法首先通过离散傅里叶变换对信号
进行特征频率的提取，再针对该特征频率进行时移傅里叶分析，通过该频率时移傅里叶分析下的时幅拐点找出突变信号的进入时刻和消失时刻，最后通过对确定时段内的信号进行二次傅里叶分析获得该突发信号的幅值． 理论分析、理论仿真和高速轴承振动数据工程仿真验证均表明，该算法能准确捕捉和提取突变信号的幅值及其在振动信号中的出现时刻和消失时刻．

As the mutation signals hidden in the vibration data of engine's high-speed bearing are difficult to pin- point in time domain,a new algorithm to obtain the amplitude-time inflection point of mutation signals is proposed.Firstly,the characteristic frequency of the signal is extracted via discrete Fourier transform.Secondly,a time-lapse Fourier analysis of the frequency is executed,which helps obtain amplitude-time inflection points.Then,the ap- pearing and disappearing moments of the mutation signal are discovered according to the inflection points.Finally,the amplitude of the mutation signal is obtained via the quadratic Fourier analysis of the signal in a certain period.The results of theoretical analysis,theoretical simulation and engineering simulation all indicate that the proposed algorithm can capture the amplitude of mutation signals accurately,and obtain the appearing and disappearing mo- ments of mutation signals in vibration data.