论文部分内容阅读
在故障诊断中常遇到以脉冲形式出现的故障信号,且伴有大小随机变化的连续噪声,甚至超过故障信号,为此,本文提出采用数理统计方法——滑动平均值及方差,将故障信号分离出来,经实验验证,该方法非常有效.
In the fault diagnosis, it often encounters the fault signal which appears in the form of pulse, and is accompanied by the continuous noise of random change in magnitude and even exceeds the fault signal. Therefore, this paper proposes using the mathematical statistics method - moving averages and variance to separate the fault signal Come out, the experimental verification, this method is very effective.