一种新的油气管道泄漏信号检测方法研究  

A New Leakage Signal Detection Technology for Oil and Gas Pipelines

作　　者：路敬祎[1,2,3] 张辉 张勇[4] 胡仲瑞[2] 李禹琦 LU Jing-yi;ZHANG Hui;ZHANG Yong;HU Zhong-rui;LI Yu-qi(SANYA Offshore Oil & Gas Research Institute,Northeast Petroleum University;Artificial Intelligence Energy Research Institute,Northeast Petroleum University;Heilongjiang Provincial Key Laboratory of Networking and Intelligent Control,Northeast Petroleum University;School of Physics and Electronic Engineering,Northeast Petroleum University)

机构地区：[1]东北石油大学三亚海洋油气研究院 [2]东北石油大学人工智能能源研究院 [3]东北石油大学黑龙江省网络化与智能控制重点实验室 [4]东北石油大学物理与电子工程学院

出　　处：《化工自动化及仪表》2024年第6期1023-1027,共5页Control and Instruments in Chemical Industry

基　　金：海南省科技专项(批准号:ZDYF2022SHFZ105)资助的课题。

摘　　要：针对管道模式识别模型效率低下、准确率不足的问题,提出一种新的管道泄漏检测方法。首先利用信号能量熵突变情况对泄漏引起的信号跳变进行有效捕捉,其次利用变分模态分解(VMD)进行去噪处理,还原信号真实动态特性。基于熵特征提取去噪信号的动态特征段,并将提取的特征向量输入改进差分进化算法(IDE)优化的支持向量机(SVM)识别模块种,实现泄漏信号的辨识。实验结果表明,与GA-SVM和PSO-SVM方法相比,IDE-SVM方法有效提高了分类识别准确率,准确率达到96.6667%。Considering low efficiency and accuracy of pipeline pattern recognition model,a new method for pipeline leakage detection was proposed.In which,having abrupt change of the signal energy entropy used to capture signal jump caused by the leakage,and the variation mode decomposition(VMD)adopted to remove noise and restore real dynamic characteristics of the signal.In addition,the dynamic feature segment of the denoised signal was extracted based on the entropy feature,including having the extracted feature vector input into the support vector machine(SVM) recognition module optimized by the improved differential evolution(IDE) algorithm to realize identification of the leakage signals.Simulation results show that,as compared to other methods,the proposed method can effectively improve system recognition efficiency along with a recognition accuracy of 96.6667%.

关 键 词：管道泄漏 能量熵 特征提取 工况识别 改进差分进化算法 

分 类 号：TQ055.81[化学工程]