燃气长直管道泄漏检测及定位方法研究
本文选题:燃气长直管道 + 泄漏检测及定位 ; 参考:《哈尔滨工业大学》2014年博士论文
【摘要】:目前,我国国民经济的持续高速发展对能源,特别是燃气资源的需求越来越迫切。随着我国西部天然气大开发和国外天然气的大量引进,燃气管道的长度越来越长,并且呈现高压力、大管径的特点。伴随着这些高压燃气管道的长度不断增加,管道泄漏事故也时有发生。因为管道内燃气压力很高,一旦发生泄漏,会造成严重的生命财产损失和环境污染。因此,及时发现泄漏并定位泄漏位置,开展应对措施,对燃气管道安全至关重要。本文采用实验和理论相结合的方法,对燃气长直管道的泄漏检测与定位做了以下的研究: 根据燃气流动基本控制方程,针对燃气管道正常工况和泄漏工况,利用特征线法分别推导了基本控制方程对应的差分方程,研究了相应的计算方法。根据燃气流动的运动方程和能量方程推导出管道沿线压力和温度分布。建立了燃气管道泄漏的储罐模型、小孔模型、大孔模型和管道模型。并利用计算算例,对模型的正确性进行了验证,实现了燃气管道泄漏量的计算。 根据实验研究目的,考虑研究对象的物理模型和实验条件,改造搭建了燃气直管道泄漏检测实验台。研究了光纤光栅应变传感器检测管道内压力变化的原理,并进行了实验测试。结合流体力学理论,对泄漏实验结果进行了分析,验证了实验研究的可行性。 为了利用基于扩展卡尔曼滤波器的方法来检测泄漏并定位泄漏位置,将滤波器估计的实时模型法的基本思想与已经建立的离散化的泄漏工况下的管道瞬态流动模型结合,,假设泄漏分布在分段点上,从而建立离散化的包含多点泄漏的瞬态流动模型。将分段点上泄漏量包含到模型中,并使其成为状态变量的一个分量。利用扩展卡尔曼滤波器对状态变量进行估计,并利用等效管道的原理,研究了实际管道泄漏量和泄漏位置的计算方法。应用模拟算例和实验数据测试验证了该方法。 负压波法是目前应用最广泛的管道泄漏检测与定位的方法,但这种方法的漏报率比较高、定位精度比较低。本文将光纤传感技术与负压波法结合,提出一种改进负压波法,并研究了此方法泄漏检测及定位的原理。研究了基于小波阈值法的负压波信号降噪的方法。通过实验对比及负压波衰减规律的研究,从技术和理论上分析了改进负压波法检测效果优于负压波法的原因。 为了应用改进负压波法定位泄漏位置,首先研究了小波分析寻找信号突变点的方法,并利用此方法得到负压波传播到上下游光纤光栅应变传感器的时间差。其次,考虑管道内气体流速变化和负压波传播速度变化对泄漏定位的影响,建立了改进的燃气管道泄漏定位算法,并利用复化辛普森和二分法进行求解。最后,应用实验数据测试验证了该方法的定位效果。 为了降低误漏报率、减少人工,将模式识别的理论引入到燃气管道的泄漏检测。提取改进负压波法采集到的泄漏工况和正常工况下的波形特征作为输入特征向量,并由此建立基于最小二乘支持向量机的燃气管道泄漏检测模型,实现了燃气管道泄漏的实时智能检测。
[Abstract]:At present, the continuous and rapid development of China's national economy is becoming more and more urgent for energy, especially gas resources. With the development of natural gas in the west of China and the introduction of foreign natural gas, the length of gas pipeline is longer and longer, and the high pressure, the special point of the large pipe diameter is accompanied by the increasing length of these high pressure gas pipelines. In addition, pipeline leakage accidents also occur. Because the gas pressure in the pipeline is very high, once the leakage occurs, it will cause serious loss of life and property and environmental pollution. Therefore, it is very important for gas pipeline safety to find leakage and locate the leakage position in time and carry out the countermeasures. This paper uses the method of combining experiment and theory to the gas. The leakage detection and location of long straight pipes have been studied as follows:
According to the basic control equation of gas flow, the differential equation corresponding to the basic control equation is derived by using the characteristic line method according to the normal and leakage conditions of gas pipeline, and the corresponding calculation method is studied. According to the motion equation and energy equation of gas flow, the distribution of pressure and temperature along the pipe is derived. A gas pipeline is established. The leaky tank model, the small hole model, the big hole model and the pipe model are used to verify the correctness of the model, and the calculation of the leakage of the gas pipeline is realized.
According to the purpose of the experiment, the physical model and the experimental conditions of the object are considered, and the test bench for the leakage detection of the gas straight pipeline is built. The principle of the change of the pressure in the pipe by the fiber grating strain sensor is studied, and the experimental test is carried out. The leakage experiment results are analyzed with the theory of fluid mechanics, and the actual results are verified. The feasibility of the study is tested.
In order to detect leakage and locate the leakage position based on the extended Calman filter, the basic idea of the real-time model method of the filter estimation is combined with the established transient flow model of the pipeline under the discrete leakage condition, assuming that the leakage is distributed on the piecewise point, thus establishing the discrete multi point leakage. The transient flow model is included in the model, and it makes it a component of the state variable. Using the extended Calman filter to estimate the state variable, and using the principle of the equivalent pipe, the calculation method of the leakage and leakage position of the actual pipeline is studied. The simulation example and experimental data test are applied. The method is proved.
Negative pressure wave method is the most widely used method of pipeline leakage detection and positioning, but the leakage rate of this method is high and the positioning accuracy is low. In this paper, an improved negative pressure wave method is proposed by combining optical fiber sensing technology with negative pressure wave method, and the principle of leakage detection and location based on this method is studied. The wavelet threshold method is studied. The method of reducing the noise of negative pressure wave signals. Through the comparison of experiment and the study of the attenuation law of negative pressure wave, the reasons for the improvement of the negative pressure wave method are analyzed from the technology and theory.
In order to locate the leakage position of the improved negative pressure wave method, the method of finding the signal abrupt point by the wavelet analysis is first studied, and the time difference between the negative pressure wave propagation to the upstream and downstream fiber Bragg grating strain sensor is obtained. Secondly, the influence of the change of gas flow velocity in the pipeline and the change of the propagation velocity of the negative pressure wave on the leakage location is considered. The improved gas pipeline leakage location algorithm and the complex Simpson and dichotomous method are used to solve the problem. Finally, the experimental data test is used to verify the positioning effect of the method.
In order to reduce the false alarm rate and reduce the manpower, the theory of pattern recognition is introduced to the leakage detection of gas pipeline. The characteristics of the leaky conditions collected by the improved negative pressure wave method and the characteristics of the normal operating conditions are extracted as the input feature vectors, and the leakage detection model based on the least square support direction quantity machine is established, and the fuel gas pipeline leakage detection model is established. Real time intelligent detection of gas pipeline leakage.
【学位授予单位】:哈尔滨工业大学
【学位级别】:博士
【学位授予年份】:2014
【分类号】:TU996.8
【参考文献】
相关期刊论文 前10条
1 陈志刚;张来斌;王朝晖;梁伟;;基于分布式光纤传感器的输气管道泄漏检测方法[J];传感器与微系统;2007年07期
2 白莉,李洪升,贾旭,崔莉,王庆国;基于瞬变流和故障检测的管线泄漏监测试验分析[J];大连理工大学学报;2005年01期
3 杨玉坤;王新;;FBG应变传感器在盾构隧道监测中应用[J];低温建筑技术;2011年02期
4 靳世久,王立宁,李健;瞬态负压波结构模式识别法原油管道泄漏检测技术[J];电子测量与仪器学报;1998年01期
5 赵廷超,黄尚廉,陈伟民;机敏土建结构中光纤传感技术的研究综述[J];重庆大学学报(自然科学版);1997年05期
6 肖智光;吴嗣跃;薛小红;;管道泄漏检测技术应用分析[J];管道技术与设备;2009年02期
7 蒲明;;中国油气管道发展现状及展望[J];国际石油经济;2009年03期
8 戚爱华;;我国油气管道运输发展现状及问题分析[J];国际石油经济;2009年12期
9 蒲明;马建国;;2010年我国油气管道新进展[J];国际石油经济;2011年03期
10 潘树新,刘耀炜;光纤传感器在水文地球化学地震前兆观测中的应用及其前景[J];国际地震动态;2001年12期
相关博士学位论文 前10条
1 杨荣根;长输石油管道泄漏检测与定位技术研究[D];南京理工大学;2011年
2 阳子轩;复杂管道泄漏检测技术研究[D];武汉理工大学;2011年
3 曲志刚;分布式光纤油气长输管道泄漏检测及预警技术研究[D];天津大学;2007年
4 胡琼;在役管线泄漏检测技术研究[D];武汉理工大学;2009年
5 孙洁娣;基于多地震动传感器的管道安全监测预警关键技术的研究[D];天津大学;2009年
6 张宇;输油管道泄漏检测新方法与关键技术研究[D];天津大学;2009年
7 邢永忠;最小二乘支持向量机的若干问题与应用研究[D];南京理工大学;2009年
8 孙良;基于模型的油气管道泄漏检测与定位方法研究[D];北京化工大学;2010年
9 许姜严;基于小波的岩石节理裂隙图像处理[D];电子科技大学;2010年
10 雷翠红;供热管网泄漏故障诊断的研究[D];哈尔滨工业大学;2010年
本文编号:2029210
本文链接:https://www.wllwen.com/guanlilunwen/chengjian/2029210.html