基于高阶累积量的齿轮系统故障检测与诊断

Fault Detection and Diagnosis of Gear System Based on Higher Order Cumnlants

详细讨论高阶累积量(Higller order cumulants,HOC)的特点和性质,分析基于HOC理论的相关信号处理技术,以及在处理随机信号的非线性特性时独特的功能。基于HOC的振动信号分析处理技术,即双谱分析、1.5维谱分析以及双相干谱分析方法,分析了各自特点以及估计算法。在此基础上,将HOC引入到机械传动齿轮系统的故障诊断分析之中,对系统故障特征进行提取并对故障类型进行区分。对不同大小和不同类型故障的传动系统进行试验测试,采用HOC进行特征提取识别,并对HOC理论实际应用效果作了分析评价,表明HOC是一种较理想有前途的有效方法。

The features and characteristics of （HOC）are discussed. Relevant signal processing technologies based on HOC, and the unique function in processing non-linear characteristics of random signals are analyzed. The vibration signal processing technologies based on HOC, such as bispectrum, 1.5-spectrum and bicoherence, are studied, and the characteristics and estimation algorithm of them are discussed. On that basis, HOC is introduced into the fault diagnosis of mechanical transmission gear systems. Tests are carried out on transmission systems with different sizes and different types of faults, and the fault features of systems are extracted and fault patterns are identified. Besides, the effect of practical application of HOC theory is evaluated, which indicates that the HOC is an ideal and effective method for the fault diagnosis of gear transmission system.