应用时间序列分析的液压溢流阀故障诊断法

基于液压溢流阀体工况,提出一种振动频率特性的故障诊断方法.利用溢流阀体振动信号的相关性、自回归模型的参数和功率谱特性,获得故障的突变信息,从而确认溢流阀的工作状态,为液压系统故障的诊断提供判断依据.测试系统采用LabVIEW虚拟仪器构建,并通过计算机自动完成测试和分析.实验结果表明,正常状态和故障状态下的电压信号没有明显的趋势,是一种平稳时间序列;有故障时,相关函数衰减比正常状态慢,但都不够明显;正常状态和故障状态下的特征根分布明显不同,故障状态下的系统可能处于某种稳定状态;正常状态的自回归(AR)功率谱比任何一种溢流阀常见故障的功率谱都低.

Fault Diagnosis of Hydraulic Relief Valve Using Time Series Analysis

In this paper,a fault diagnosis method for hydraulic relief valve has been put forward according to their vibration frequency characteristics.Using the correlation analysis,the parameters and spectra of a time series model of the vibrant signals from the relief valve,the abrupt change informations can be obtained.Therefore the fault conditions of the relief valve can be confirmed and the decision support is provided for the fault diagnosis of hydraulic relief valve.The experiment and diagnosis process based on virtual instruments LabVIEW are automatically controlled by computer in this paper.The experiment shows that voltage signals on normal and fault states are a steady time series which have no apparent tendency;the correlated functions on fault states attenuate more slowly than one on normal state,but they are not pretty apparent;the distributions of characteristic roots are obviously different between normal and fault states;the system on fault states may be in a stable condition;auto-regressive(AR) power spectrum curve on normal state is different from other fault states of hydraulic relief valve.