基于光纤应变测量技术的管道泄漏监测及定位研究

Pipeline transportation of liquids and gas is a vital part of the national economy. As the contents of the fluids inside pipelines are harmful to humans and the environment, there is a priority placed on developing methods for detecting damage, such as leakage, in pipelines. Over the past years, a variety of leakage detection methods have been developed due to the emergence of new materials and sensor technologies. This project presents a quasi-distributed fiber strain sensing based approach to pipeline leakage monitoring and localization. Based on the negative pressure wave(NPW) phenomenon induced by leakage, the strain distribution along a pipeline is acquired to determine the leakage point. The project starts from a research of leakage mechanism and intends to propose several leakage localization methods for different leakage cases, including NPW arrival time difference method, NPW energy attenuation method and steady state strain distribution identification method. Another branch of the project is the development of fiber Bragg grating(FBG) sensor with higher strain sensitivity, which determines the performance of the leakage monitoring system for micro leakage rate. This research will investigate the performance of the novel FBG sensor, including the sensitivity, stability, etc. and then demonstrates its effectiveness and practicability in pipeline leakage monitoring. Combined with the pipeline leakage monitoring software planned to develop, all the mentioned localization methods and sensing technique will be verified by pipeline model tests. Furthermore the minimum leakage rate required for detection by the method can be significantly reduced by increasing the number of hoop strain measurement points. In this research direction, an analytical model of the NPW energy attenuation was used to determine the best sensor spacing along a pipeline according to different leakage rate requirements. A potential researching topic in this project is the development of smart pipeline structure by integrating fiber sensing technique.

管道泄漏监测技术对于及时了解管道安全运营状态，确保国民经济和人民生命财产安全具有重要意义。随着新型材料、新型传感器的不断发展，各种管道泄漏检测及监测技术应运而生。本项目提出了一种基于准分布式光纤应变测量技术的管道泄漏监测及定位的方法，它以负压波传播规律为基础，通过采集泄漏引起的管道沿线的应变分布信息，对管道泄漏位置进行判断。本项目将从管道泄漏的机理研究出发，给出不同泄漏情况下的泄漏定位方法，包括负压波时间差定位法，负压波能量衰减定位法及稳态应变分布模式识别法等。而高灵敏度光纤光栅传感器的研发也是本项目的重要研究分支，它决定了泄漏监测系统对于微小泄漏量的检测能力。结合本项目拟开发的管道泄漏监测软件系统，上述定位方法及传感技术均将通过管道泄漏模型试验进行验证。利用光纤传感技术开发智能管道结构也将是本项目一个潜在的技术研究工作。

管道是国家重要公共基础设施，其泄漏监测具有重要意义。本项目提出了基于分布式光纤应变测量信号，进行管道泄漏事故判别及泄漏定位的新技术，目前已形成了“泄漏机理→传感技术→监测算法→系统设计”的较为完整的管道泄漏监测理论和技术体系。在泄漏监测原理方面，提出利用特征线法分析管道泄漏稳态压力分布规律，结合机器学习模型计算管道泄漏位置的方法。在传感技术方面，针对前期开发的光纤光栅（FBG）管道泄漏传感器，进行结构及参数优化，提升传感器对管道泄漏监测可检泄漏量的分辨能力。在泄漏监测算法方面，针对管道泄漏事故原始数据不易获取的问题，提出利用适于小样本学习的支持向量回归（SVR）网络模型，分析管道泄漏位置与光纤应变分布的复杂非线性映射关系，并利用多种优化方法获得更高的管道泄漏定位准确率。搭建管道泄漏监测模拟试验平台，对上述相关监测技术及监测系统进行模拟试验测试。经理论和试验验证，本项目所提出的基于分布式光纤应变测量的管道泄漏监测技术，对于管道泄漏事故识别准确率高达97%。通过对真实长输管道的模拟计算，该方法对于管道沿线各泄漏点定位平均误差率低于1%，且系统具有较好的鲁棒性。同时，通过改进光纤传感器泄漏监测灵敏度，可将泄漏检测系统在管道最小可检泄漏率方面检测性能提升超过30%。本项目所研究的管道泄漏监测及定位技术，有望应用于长输油气管道、城市管网、海底管道等各类管道安全监测系统中，具有潜在的实际应用价值。