基于混沌特性和递归图的输油管道泄漏检测与定位方法

发布时间：2018-01-26 11:23

本文关键词： 输油管道泄漏检测混沌特性关联维双树复小波变换递归图　出处：《燕山大学》2015年硕士论文　论文类型：学位论文

【摘要】：近年来,油气产品已经对我们的生活越来越重要。随着我国管道事业的发展,部分输油管道使用的时间越来越长,施工过程的缺陷以及自然腐蚀等原因,造成输油管道发生泄漏,特别是人为的破坏、盗油等行为更是给人民的安全、国家的财产、自然环境带来了很大的威胁。因此当输油管道发生泄漏时,能够及时的发现泄漏并确定泄漏点位置变得越来越重要。论文阐述了输油管道泄漏检测与定位的研究现状,分析了常用的输油管道泄漏检测与定位方法。针对单一参数研究输油管道泄漏可能引起误报警或是定位准确率较差等缺点,论文提出基于数据采集与监控系统(SCADA),以流量及压力双参数为处理对象,引入混沌特性分析及递归图方法对管道泄漏检测与定位方法进行分析,主要内容包括:首先分析了输油管道泄漏时流量及压力的变化情况,提出以流量及压力双参数为处理对象,采用混沌特性分析及递归图方法实现管道泄漏检测与定位的方案。给出负压波定位公式,并在实现定位过程中对负压波波速进行了改进。然后以输油管道进站口流量信号为处理对象,基于混沌特性分析识别管道泄漏情况。首先对系统动力学特性进行相空间重构,利用微分熵法同时求取延迟时间和嵌入维数,避免了分开求取的不一致性,然后采用Wolf算法求取Lyapunov指数,确定系统的混沌特性,最后通过GP算法计算关联维数,分析系统的混沌特性。最后根据SCADA系统采集到的负压波信号,采用双树复小波变换对其进行处理,提取特征信号,判断输油管道是否发生泄漏,然后利用递归图平均灰度法进行特征点定位,确定特征时刻,根据另一端确定的特征时刻即可确定负压波传输到首末两端的时间差,最后结合定位公式及负压波波速确定泄漏点具体位置。
[Abstract]:In recent years, oil and gas products have become more and more important to our daily life. With the development of pipeline industry in China, some oil pipelines have been used longer and longer, the defects in construction process and natural corrosion and other reasons. Oil pipeline leakage, especially man-made damage, oil theft and other acts to the safety of the people, national property, natural environment has brought a great threat. It is becoming more and more important to find leakage and determine the location of leak point in time. This paper describes the current situation of oil pipeline leak detection and location. The common leak detection and location methods of oil pipeline are analyzed. According to the single parameter, the leakage of oil pipeline may cause false alarm or poor location accuracy and so on. In this paper, based on the data acquisition and monitoring system (SCADAA), the flow and pressure parameters are taken as the processing objects, and the chaos characteristic analysis and recursive graph method are introduced to analyze the pipeline leakage detection and location method. The main contents are as follows: firstly, the variation of flow rate and pressure during oil pipeline leakage is analyzed, and the double parameters of flow and pressure are proposed as the processing object. The scheme of pipeline leakage detection and location is realized by using chaos characteristic analysis and recursive diagram method. The negative pressure wave localization formula is given. The negative pressure wave velocity is improved in the process of positioning. Then the flow signal at the inlet of oil pipeline is taken as the processing object. Firstly, the phase space reconstruction of system dynamics is carried out, and the delay time and embedding dimension are obtained simultaneously by differential entropy method, which avoids the inconsistency of separation. Then the Wolf algorithm is used to calculate the Lyapunov exponent, and the chaos characteristic of the system is determined. Finally, the correlation dimension is calculated by GP algorithm. Finally, according to the negative pressure wave signal collected by the SCADA system, the double tree complex wavelet transform is used to process the chaotic characteristics of the system, to extract the characteristic signal, and to judge whether the oil pipeline leak or not. Then the recursive image average gray level method is used to locate the feature points, and the time difference between the two ends of the negative pressure wave transmission can be determined according to the characteristic time of the other end. Finally, combined with the localization formula and the wave velocity of negative pressure, the specific location of leak point is determined.
【学位授予单位】：燕山大学
【学位级别】：硕士
【学位授予年份】：2015
【分类号】：TE973.6

【参考文献】