基于EMD的油气管道安全分布式光纤预警系统信号分析方法

针对油气管道安全分布式光纤预警系统检测信号的非平稳特征，提出了基于经验模态分解（EMD）的信号分析方法．预警系统基于Mach—Zehnder光纤干涉仪原理，沿油气管道同沟敷设光缆，利用其中三条单膜光纤构成分布式微振动测试传感器，实时检测管道沿途所发生的泄漏及其他异常事件．该信号分析方法将检测信号通过EMD分解为多个平稳的固有模态函数（IMF）之和，选择若干包含信号主要信息的IMF分量对其进行峭度分析，提取几种事件信号的特征．最后分析了现场实验数据及其信号处理结果．研究结果表明，该方法可以有效地对管道周围发生的泄漏和其他异常情况进行特征提取．

EMD-Based Signal Analysis Method of Distributed Optical Fiber Pre-Warning System for Safety of Oil and Gas Pipeline

A signal analysis method based on empirical mode decomposition(EMD) is proposed considering at the non-stationary characteristic of the detected signal by distributed optical fiber pre-warning system, for safety of oil and gas pipelines. In the pre-warning system, based on the principle of Mach-Zehnder optical fiber inter-ferometer , an optical cable is laid along the pipeline in the same ditch, and three of the single-mode optical fi-bers in the cable are used to build up the distributed micro-vibrant measuring sensor, by which the real-time de-tection of leakage and other abnormal events can be realized. In this signal analysis method the detected signal can be decomposed into a sum of finite stationary intrinsic mode functions (IMF) by EMD, among which some IMF components containing the main information of signals are selected to be analyzed on its kurtosis. Therefore the eigenvectors of the vibrant signals led by leakage or other abnormal events can be extracted. Finally, experi-mental data obtained at oil field and the results of the signal processing are analyzed. The analysis shows that the eigenvectors of vibrant signals led by leakage and abnormal events can be extracted by the signal analysis method based on EMD efficiently.