基于LMD-STA/LTA模型的油气管道泄漏检测方法

Pipeline leak detection method based on LMD-STA/LTA model

为了降低流体混杂噪声和传感器漂移对负压波信号的影响,提高负压波法油气管道泄漏检测的定位精度,采用局部均值分解法(LMD)对压力信号进行分解,对降噪后的信号重构后,采用长短时窗比法(STA/LTA)确定上下游压力传感器接收到的信号时差,并通过实际放油试验完成泄漏检测与定位。结果表明,该方法可有效识别突变的负压波信号,可识别1%~2%泄漏量的快速泄漏和缓慢泄漏,有效缩小了泄漏点的查找范围。研究结果可为油气管道的监测与预警提供参考。

In order to reduce the impact of noise on negative pressure wave signal,improve the positioning accuracy of negative pressure wave,the pressure signal is decomposed by local mean decomposition method,and the time difference between upstream and downstream sensors is determined by STA/LTA method with the signal reconstructed based on noise reduction,and the leakage detection and location are completed through the actual oil discharge test.The results show that the method can effectively identify the abrupt negative pressure wave signal,and can identify the fast leakage and slow leakage of 1%-2%leakage volume,which effectively reduces the search range of the leakage point.The research results can provide practical reference for oil and gas pipeline monitoring and early warning.