论文部分内容阅读
随机振动台根据功率谱密度(PSD,Power Spectral Density)生成的信号为高斯信号,而实际振动环境有时是非高斯的,因此随机振动实验常常无法准确模拟产品在真实振动环境下的失效情况.通过两个案例分别对均方根值(RMS,Root Mean Square)不随时间变化和随时间变化的非高斯随机振动进行了模拟方法研究.案例1利用Hermite多项式法对高斯信号进行了转换,在保证功率谱密度不变的同时得到了具有指定峭度的RMS不随时间变化的非高斯信号,但该方法对于输入的峭度有限制,当输入峭度大于10时,误差达到了20%.案例2利用一种新方法对实测的RMS随时间变化的非高斯振动进行了模拟,模拟后得到的非高斯信号和实测信号具有相同的功率谱密度、峭度以及概率分布,验证了新方法的准确性.
The random vibration table can not exactly simulate the failure condition of the product in the real vibration environment because the signal generated by random power table based on PSD (Power Spectral Density) is Gaussian signal, but the actual vibration environment is sometimes non-Gaussian. In this case, we study the non-Gaussian random vibration whose root mean square (RMS) does not change with time and change with time respectively. Case 1 The Gaussian signal is converted by Hermite polynomial method, With constant density, non-Gaussian RMS signals with specified kurtosis do not change over time, but this method has a limit on the kurtosis of the input, with an error of 20% when the input kurtosis is greater than 10. Case 2 Using a The new method is used to simulate the measured non-Gaussian vibration of RMS with time. The simulated non-Gaussian signal and the measured signal have the same power spectral density, kurtosis and probability distribution, which verifies the accuracy of the new method.