摘要
针对穿跨越油气管道人工巡检困难的问题,文中采用分布式光纤实时监测管道状态,对振动信号进行分类,识别管道振动源。提出改进的秃鹰搜索算法和光纤振动信号特征提取方法,并基于神经网络对分布式光纤相位敏感信号进行分类和识别。实验结果显示,文中提出的Ct-GBES-BPNN分类模型具有良好的分类识别效果,可为保障穿跨越油气管道安全提供支撑。
Addressing the challenge of manual inspection difficulties for trans-crossing oil and gas pipelines,this paper employs distributed fiber optics for real-time monitoring of pipeline conditions,classifying vibration signals to identify the sources of pipeline vibrations.An improved Bald Eagle Search algorithm and a method for extracting features from fiber optic vibration signals are proposed.Furthermore,distributed fiber optic phase-sensitive signals are classified and recognized based on neural networks.Experimental results demonstrate that the proposed Ct-GBES-BPNN classification model achieves excellent performance in classification and recognition,providing support for ensuring the safety of trans-crossing oil and gas pipelines.
作者
徐彩军
王芳
张世杰
于漫漫
XU Caijun;WANG Fang;ZHANG Shijie;YU Manman(Fujian Boiler and Pressure Vessel Inspection Institute,Fuzhou 350008,Fujian,China;Fuzhou Huarun Gas Co.,Ltd.Fuzhou 350001,Fujian,China)
出处
《市场监管与质量技术研究》
2024年第1期23-27,46,共6页
Market Regulation and Quality Technology Research
基金
国家市场监督管理总局技术保障专项(2022YJ17)
福建省科技厅引导性项目(2022H0032)。
关键词
分布式光纤
相位敏感信号
油气管道
特征提取
振动识别
Distributed fiber optics
Phase-sensitive signals
Oil and gas pipelines
Feature extraction
Vibration recognition
