EEMD-四线性平行因子的机械多故障欠定盲分离方法

Blind Separation Method for Mechanical Faults Based on Ensemble Empirical Mode Decomposition and Quadri-linear Parallel Factors

将四线性平行因子和集合经验模态分解(Ensemble Empirical Mode Decomposition,EEMD)结合,提出基于EEMD的四线性平行因子欠定盲源分离方法。在所提出的方法中,利用EEMD方法分解观测信号,得到各子信号分量,然后选择有效的相关子分量信号并将相关子分量信号与原观测信号进行重新组合,得到新的观测信号,解决欠定盲分离问题。然后,将新观测信号利用四线性交替最小二乘法迭代(Quadrilinear Alternating Least Squares,QALS)进行拟合得到载荷矩阵估计和混合矩阵估计,用最短路径法得到源信号的估计。仿真和实验结果验证了所提方法的有效性。

An under-determinate blind source separation method based on ensemble empirical mode decomposition(EEMD) and quadri-linear parallel factors is proposed.In this method,the observation signal is decomposed by EEMD method to obtain several subcomponents.Then the effective correlation subcomponent signals are selected to reconstruct a new observation signal with the original observation signal.Thus,the problem of the under-determinate blind separation is transformed into a positive or over-determinate blind separation problem.Then,the new observation signal is fitted by the quadri-linear alternating least squares iterative method(QALS) to obtain the load matrix estimation and mixed matrix estimation.The estimation of source signal is obtained by the shortest path method.The simulation and experimental results show that the proposed method is very effective.