不同管路元件流动声发射信号熵值量化分析方法

Quantitative Analysis Method for Acoustic Emission Signal Entropy of Fluid Flow in Different Pipeline Components

针对流体在流经不同管路元件时产生的噪声会造成管道损伤在线监测精准性下降的问题,引入了小波能量熵与时频熵两种熵值量化分析方法,通过计算流体流动声信号的熵值对流动过程中不同管路元件的流动状态进行量化,从而判断损伤是否发生以及损伤所处的管路元件部位。在此基础上,设计了管路流动声发射测试实验,采集了不同压力下直管、弯头入口、弯头出口、三通、变径管小头共5种管路元件的声发射信号,对所采集的声发射信号逐一采用时频熵及小波能量熵的方法分析计算。测试实验结果表明,时频熵与小波能量熵均能很好地区分不同管路元件的流动状态差异,且时频熵对同一管路元件压力变化时的流动状态有着较高的量化精度。

To address the problem that the noise generated by the fluid flowing through different pipeline components would cause the accuracy of online monitoring of pipeline damage to decrease,two entropy quantitative analysis methods,wavelet energy entropy and time-frequency entropy were introduced.The entropy value of the acoustic signal of fluid flow was calculated to quantify the flow state of different pipeline components during the flow process,so as to determine whether the damage had occurred and the pipeline component where the damage was located.On this basis,pipeline flow acoustic emission test experiment was designed,and the acoustic emission signals of five types of pipeline components were collected under different pressure:straight pipe,elbow inlet,elbow outlet,tee and reducer head,and the collected acoustic emission signals are analyzed and calculated by the method of time-frequency entropy and wavelet energy entropy one by one.The test results show that the time-frequency entropy and wavelet energy entropy can distinguish the flow state differences of different pipeline components,and the time-frequency entropy has a high quantification accuracy for the flow state when the pressure of the same pipeline component changes.