一种新的非高斯随机振动信号的模拟方法

The traditional environmental vibration tests usually assume that the random vibration signal follows Gaussian distribution and power spectral density (PSD) is used as the test conditions, while in practice, lots of vibration excitations to the structure are non-Gaussian distribution and PSD can only describe the low-order statistics of signals but not for the high-order statistics of the non-Gaussian signals such as the kurtosis and skewness. So a method based on amplitude modulation technique (AMT) of window function to generate the non-Gaussian signal with PSD and kurtosis is developed. An approximated simulation method is provided to generate the modulation signal. Two kinds of statistical distribution, Beta and Weibull, are used to construct modulation signal, the relation between target kurtosis and modulation signal distribution parameters is explored, and the suitable kurtosis range of two distributions is discussed. A case study is presented to show that the synthesized non-Gaussian signal has the same PSD, probability density function (PDF), and kurtosis as outfield measured data, which verifies the correctness of the method.