【摘要】 基于高阶谱能够抑制高斯信号,并且可以在较强的背景噪声中提取故障信息的特点,在分析高阶累积量与高阶谱的理论基础上,针对牙轮钻头运转中振动信号的非线性、频谱成分多样性和二次相位耦合性等特点,研究了基于高阶谱分析的机械故障特征提取方法,提出了基于双谱估计的钻头故障诊断方法。试验结果表明,双谱估计方法不仅能有效地区分出正常牙轮和故障牙轮,也能很好地区分出牙轮不同的故障类型;能有效地抑制高斯噪声,可以在较强的背景信号中有效地提取故障信号;可作为牙轮钻头早期故障诊断的量化指标。更多还原

【Abstract】 Higher-order spectrum is able to suppress Gaussian signal and extract failure information in relatively strong background noise.Considering this feature and based on the theory of high-order cumulant and high-order spectrum,a study was conducted of the method to collect mechanical failure features,and a bi-spectrum estimation-based diagnosis method for roller bit failure was formulated on the basis of high-order analysis and by considering the non-linearity of vibrating signals,diversity of spectral elements and coupling of quadratic phase in the rotation of roller bit.The test results show that the bi-spectrum estimation method can not only distinguish normal roller cone from broken-down roller cone,but also identify different kinds of roller failure.It can effectively inhibit Gaussian noise and effectively collect failure signal in strong background signals.It can serve as a quantitative index in early-stage failure diagnosis of roller bit.更多还原