长距离输油气管道泄漏监测与准实时检测技术研究

Study on Leakage Monitoring and Quasi-Real-Time Detection Technology for Long Distance Oil and Gas Pipeline

在实验研究基础上,采用巴特沃兹滤波方法去除信号中的噪声干扰,将基于傅立叶方法的故障诊断方法引入长距离输油气管道泄漏监测与准实时检测技术中,对可以反映油气的泄漏压力信号的相关特征向量指标进行了系统研究。该方法对实验得到的压力信号进行快速傅里叶变换,解决了对时域信号向频域转化的难题,进而通过对加窗后重构的新序列进一步分析,就能根据不同频域内的特征向量值完成对产生泄露原因的识别。并在后期的实际应用中取得了相对较好的效果,同时在上述研究基础上,对总结油气管道泄露检测与准实时检测技术进行分析,为该技术的进一步发展与完善提供指导性建议。

On the basis of experimental research,the Butterworth filtering method is used to remove the noise interference in the signal.The fault diagnosis method based on Fourier method is introduced into the leak monitoring and quasi-real-time detection technology of long-distance oil and gas pipelines,and the related eigenvector indexes reflecting the leak pressure signal of oil and gas are studied.Unified research.This method makes fast Fourier transform for pressure signals obtained from experiments,and solves the problem of transforming timedomain signals into frequency-domain signals.By further analyzing the new sequence reconstructed after adding windows,we can identify the causes of leakage according to the eigenvectors in different frequency domains.At the same time,on the basis of the above research,the technology of leak detection and quasi-real-time detection of oil and gas pipelines is summarized and analyzed,which provides guidance for the further development and improvement of the technology.