摘要
为保障埋地燃气管道的安全运行,提出基于光纤传感技术的埋地燃气管道预警系统。系统由光源模块、传感控制模块、信号处理模块和预警模块等结构组成,利用光纤光栅应力传感器采集可能对管道造成破坏的行为所产生的土壤振动信号,将信号时频特征、能量窗口特征和频域特征合并为振动信号特征向量,分析其在不同危害程度下的光纤预警响应规律;利用支持向量机与BP神经网络构建分类器,依照特征向量区分并确定信号安全与否,利用预警模块对非安全信号发出预警信息;结合GIS与SCADA系统等技术构建应急救援指挥决策模块,提升光纤预警系统的智能化程度。实验结果表明,该方法可准确识别不同性质的破坏行为,平均识别分类精度高达92.64%,且预警定位误差在300 m之内,预警响应时间低于300 ms,可有效实现管道危险预警。
In order to guarantee the safe operation of buried gas pipeline,an early warning system based on optical fiber sensing technology is proposed.System is controlled by a light source module,sensor module,a signal processing module and the warning module structure,such as the use of fiber Bragg grating stress sensor acquisition can damage behavior of the pipes produced by the soil vibration signal,the signal time-frequency features,characteristics and frequency domain energy window into vibration signal feature vector,to analyze its optical fiber under different damage degree early warning and response;The classifier is constructed by using SVM and BP neural network,and the signal security is distinguished and determined according to the feature vector.Combined with GIS and SCADA system and other technologies,the emergency rescue command and decision module is constructed to improve the intelligence of the optical fiber warning system.The experimental results show that this method can accurately identify the damage behaviors of different properties,with an average recognition and classification accuracy as high as 92.64%.Moreover,the early-warning positioning error is within 300 m,and the early-warning response time is less than 300 ms,which can effectively realize the early-warning of pipeline danger.
作者
梁金禄
黄小玉
杨军
谢廷远
陈家雄
方丽萍
LIANG Jinlu;HUANG Xioayu;YANG Jun;XIE Tingyuan;CHEN Jiaxiong;FANG Liping(College of Petroleum and Chemical Engineering,Beibu Gulf University,Qinzhou Guangxi 535011,China)
出处
《激光杂志》
北大核心
2020年第3期139-143,共5页
Laser Journal
基金
广西科技重点研发计划项目(No.桂科AB16380285)。
关键词
光纤传感技术
埋地燃气管道
预警
振动信号
分类
optical fiber sensing technology
buried gas pipelines
early warning
vibration signal
classification