盲源分离方法在旋转机械故障诊断中的应用

在旋转机械故障诊断中,从被监测机器测得的振动信号除噪声干扰外,还常混叠有其它机器的振动。由所测得的此混叠信号中分离出被监测机器的信号,并降低噪声的影响,是提高故障诊断准确性的基础。这里采用四阶累积量迫零算法对机器振动信号进行分离,通过仿真信号及实验室实测振动信号的分离验证,此方法是有效的,可作为旋转机械故障诊断的信号预处理方法。

Application of blind source separation in rotating machine diagnosis

In implementing the fault diagnosing of rotating machine, the measured vibration signals, include both noise and the signals from other machines. Therefore it is an essential step to separate the individual signals of each machine and reduce noise from the signal mixture . In this paper, the source separati on is performed by means of fourth order cumulants methods, with which the vibra tion signals will be separated. The computer simulation and experimental verific ation show that the algorithm can well restore source signals from their mixture ..This method can be seen as a pre-processing step that improves the diagnosis.