涡街流量计信号的高阶统计量特征

Higher-Order Statistics Characteristics of Vortex Flowmeter Signals

涡街信号的正确分析和有效处理是实现涡街流量计准确测量的先决条件之一.为研究涡街信号的非高斯性,提出了一种基于高阶统计量方法的涡街流量计信号分析的新方法.通过对涡街管壁差压信号的双谱分析,提取了不同流量下信号的非高斯特征.以信号的双谱幅度最大值作为特征参数,对不同流量下涡街信号偏离高斯分布的程度进行了定量估计.实验结果表明:涡街信号的非高斯性随着流量的增大而不断增加.这一研究结果为深入理解涡街现象、进一步优化涡街流量计的设计提供了有力的参考.

Correct analysis and effective processing of vortex signals is one of the prerequisites for accurate use of vortex flowmeters. A novel method based on higher-order statistics analysis was proposed to investigate the non-Gaussian characteristics of vortex flowmeter signals. The non-Gaussian characteristics of vortex signals at various flowrates were extracted by the bispectral analysis of duct-wall differential pressures from the vortex flowmeter. The maximal values of bispectra of vortex signals were taken as eigenvalues for the quantitative estimates of the deviation degree from the normality. The experimental results show that the non-Gaussian degree of vortex signals gets increased with the increment of flowrates. These findings give a helpful reference for people to get a more in-depth understanding of vortex phenomena and to improve the design of vortex flowmeters further.