目前使用的大多数盲源分离方法都依赖于观测传感器数量等于信号源数目这样一个基本假设。主要针对传感器数量大于源信号数量情况,对涡扇发动机振动信号盲源分离问题进行研究,先后采用三种算法进行试验,结果证明了奇异值降维方法和基于四阶累积量的源数估计方法的可行性和可靠性。
Most of the blind source separation methods were dependent on the fact that the observing sensors'number are equal to the number of signal sources.This paper is aimed to research into the blind source separation of Vibration Signal Processing from turbofan engine when sensors number are more than signal sources'number.Three methods are adopted and signal simulation in experiment shows the ways based on Singular Value Decomposition and the method based on fourth-order cumulants are feasible and reliable.