基于负熵最大化的机械振源半盲分离方法

Semi-blind sources separation of mechanical vibrations base on maximization of negentropy

为了快速、有效地分离传感器观测信号中的机械振源信号,提出一种负熵最大化的机械振源半盲分离方法.该方法根据目标振源的振动特性构造相应的参考源信号,将参考源信号和分离的目标振源信号的均方误差作为约束条件引入到盲源分离的对照函数中,通过求解约束最优问题,实现目标机械振源信号的分离.试验结果表明,基于负熵最大化的半盲分离方法能快速、有效地分离出目标振源信号,为机械振动信号的监测与故障诊断提供一种新的方法和思路.

In order to separate mechanical vibration sources form sensor signals rapidly and effectively, a novel method was proposed, which was based on maximization of negentropy for semi-blind sources separation of mechanical vibrations. The reference signals that carry some information of sources were constructed. The mean square error between reference signals and separated sources was incorporated into contrast function as the constraints. The interested mechanical vibration source was obtained by solving the constrained optimization problem. The proposed method was compared with the conventional BSS method, and the experiment results showed that the proposed method is very effective. It is possible to apply the new method to vibration signals analysis and mechanical fault diagnosis.