金第1期
第38卷
2017年1月
仪器仪表学报 Chinese Journal of Scientific Instrument
Vol. 38 No. 1 Jan.2017
基于动态核独立元统计量的石油管道泄漏检测
率小花"2，李景哲”，李太福,唐海红，刘兴华
（1.重庆科技学院电气与信息工程学院重庆401331;2.四川理工学院四川省人工智能重点实验室自贡
643000）
商要：石油管道泄漏是受腐蚀、磨损、焊缝缺陷、振动、冲刷以及人为破坏等多种因素影响的连续动态过程，单纯基于压力信号摘
的检测和基于高斯分布假设的信号分析方法不能适应其多变量、强耦合、动态特性。为此，综合考虑与管道泄漏有关的操作参数和环境参数，针对管道监测参数呈现时序自相关性、泄漏检测精度不高的间题，提出一种基于动态核独立分量分析（DKICA）的石油管道泄漏检测方法。首先引人动态特性确定算法(DOD)计算模型最佳参数阶次，解决动态过程导致的监测参数呈现时序自相关性问题；再采用核独立分量分析（KICA)在核主元空间提取独立元；最后通过考察独立元的T、SPE联合指标判断泄漏发生。通过对某一输送场站采集的数据进行实验验证，结果表明采用联合指标D，的正常样本误检率和泄漏样本漏检率都远低于单独采用T或SPE统计量；而引人动态特性的2阶DKICA对于正常样本的误检率和泄漏样本的漏检率都低于未引人动态特性的KICA方法。可见，所提出的基于动态核独立分量联合指标的石油管道泄漏检测方法是一种高效且可行的方法。关键词：泄漏检测动态核独立分量分析；联合指标；石油管道；计算机模拟
中图分类号：TP183
937
TH865
文献标识码：A
国家标准学科分类代码：510.4010
Oilpipelineleakdetectionbasedonindexofdynamic
Kernelindependentcomponents
Gu Xiaohua'2,Li Jingzhe',Li Taifu',Tang Haihong' ,Liu Xinghua
(1. College of Electronic & Information Engineering, Chongging University of Science and Technology, Chongging 401331,China 2. Artificial Intelligence of Key Laboratory of Sichuan Province, Sichuan University of Science & Engineering, Zigong 643000, China)
Abstract: Oil pipeline leak is a continuous and dynamic process affected by many factors (e. g. , corrosion, wear, weld defects, vibration, erosion and man-made destruction). The method based on pressure signal detection and Gaussian assumption signal analysis cannot meel the characteristics of multivariable, strong coupling and dynamics, In this article, the operating and environmental parameters associated with the pipeline leaking are comprehensively considered. A novel oil pipeline leak detection method based on Dynamic Kernel Independent Component Analysis (DKICA) is proposed to solve the timing-sequence-autocorrelation problem of the pipeline monitoring parameters and enhance the detection accuracy. Firstly, the optimal order of the model parameters is confirmed by the determination characteristies of dynamic (DOD) algorithm to reduce the autocorrelation among the monitoring parameters. Secondly, the Kemel Independent Component Analysis (KICA) is utilized to extract the independent component in kemel principal space. Finally, the pipeline leak is monitored by T°, SPE and the combined index of the independent components. Experimental results indicate that both the missing and false detection accuracies of the combined index D, are much lower than thoee of the SPE and T° separately. Additionally, both the missing and false detection accuracies of the 2-order DKICA are much lower than those of KICA, due to the consideration of the dynamic characteristics. It verifies the feasibility and effectiveness of the proposed method based on DKICA for the oil pipeline leak detection.
Keywords:leak detection; dynamic Kermel independent analysis; combined indices; oil pipeline; computer simulation
收稿日期：2016-07
Received Date: 2016-07
*基金项目：国家科技重大专项（2016ZX05017004）、国家自然科学基金（51404051）重庆市基础科学与前沿技术研究（一般）项目（cestc2015jeyjA90024）、重庆市基础科学与前沿技术研究（重点）项目（cstc2015jeyjBX0089）、重庆市教委科学技术研究项目（KJ1401312，
KJ1501304)）、重庆科技学院校内基金（CK2016Z16)项目资助万方数据