基于时域随机化的超高斯真随机驱动信号生成技术研究  

Study on the generation of super-Gaussian and true-random drive signals using time domain randomization

作　　者：蒋瑜[1] 陈循[1] 陶俊勇[1] 

机构地区：[1]国防科技大学机电工程与自动化学院,湖南长沙410073

出　　处：《振动工程学报》2005年第4期491-494,共4页Journal of Vibration Engineering

基　　金："十五"国家部委重点预研项目资助(41319030101)

摘　　要：提高振动模拟试验的真实性和增强振动激发试验的效力都需要产生超高斯分布的真随机驱动信号。研究了基于时域随机化的超高斯真随机驱动信号生成技术,针对常用半正弦窗函数和不同重叠因子值进行了理论分析和数值仿真,结果表明:时域随机化前后的伪随机和真随机信号峭度值是线性关系,但输出的超高斯真随机信号峭度值比输入的超高斯伪随机信号峭度值要小,并且重叠因子取值越大峭度值减小的程度越大。To enhance the reality of vibration simulation testing and the effectiveness of vibration stimulation testing, it is necessary to generate true random drive signals with super-Gaussian distribution. This paper studies the technique of generating super-Gaussian and true-random drive signals based on time domain randomization. The results of theoretical analysis and numerical simulation for half-sine window function and different overlap factor values show that the relationship between the kurtosis values of input and output is linear, but the output kurtosis value is less than the input and decreases with the increasing overlap factor value.

关 键 词：随机振动 试验 时域随机化 超高斯 峭度 

分 类 号：TB535[理学—物理] TB311.5[理学—声学]