振动源信号的快速二阶统计量算法研究

RAPID SECOND-ORDER BLIND SOURCE SEPARATION ALGORITHM FOR VIBRATION SIGNALS

快速而有效地分离振动源混合信号,是故障监测系统进一步诊断的基础。提出了一种改进的基于二阶统计量的盲源分离算法。算法采用平均时滞相关矩阵,以获取混合矩阵的平均特征结构,实现信号的有效分离。仿真和实验结果表明,该方法具有计算简单,运算速度快的特点,为振动信号在线监测与故障诊断提供了一种新的方法。

A rapid second-order blind identification algorithm was developed to separate temporally correlated sources with different normalized spectra based on the original algorithm of blind source separation. The set of convariance matrices of the original algorithm was substituted by their average matrix to obtain their' average eigenstructure'. And the problem of joint diagonalization of several matrices was reduced to one matrix diagonalization. So the cost of the computation was reduced. In contrast to the original algorithm, the modified algorithm enhances the speed of computation obviously without leading to the degeneration of separation performance. The numerical simulations and real experiments illustrate the effectiveness and rapidness of the proposed method. It is possible to apply the method to real-time signal analysis due to its rapidness and efficiency.