天然气长输管道泄漏点的检测与定位方法研究
|
摘要
作为五大运输行业之一的管道运输系统，以其环保、经济、快捷等优点，在国民经济中的地位越来越为突出。随着我国西部开发战略的实施，必将有大量的油气管道投入建设和运行。然而近些年来，由于管道的自然寿命和人为因素等的影响，管道泄漏事故频繁发生，不仅造成大量物质损失和环境污染，还带来了重大的人身伤亡事故。及时发现管道泄漏和精确确定泄漏点位置对于减少损失、维护管道安全运行显得十分重要。为将泄漏事故造成的各种危害减少到最小，需要研究泄漏检测技术，以达到更高的泄漏检测灵敏度和更准确的泄漏点定位精度。
鉴于天然气长输管道的模型受到流体特性、地势状况及管道自身的特性等诸多复杂因素的制约，难于建立其精确的数学模型，新铺设的输气管网均已配有SCADA系统，对现场各类数据有了丰富的储备等因素，同时结合我国管道输送的实际情况，并借鉴目前国际上各种泄漏检测与定位的先进方法，针对天然气长输管线泄漏故障的特点，本文提出了基于小波分析和负压波相结合的检测泄漏与定位技术。其中，主要进行了以下几方面的工作：
针对泄漏产生的负压波在管道中的传播速度受媒介的密度、压力、比热及材质等因素的影响，而非声波在空气中传播的速度，本文利用能量守恒原理，对其进行了修正；考虑到管内气体流速对压力波速的影响，对原始的负压波定位公式进行了一定程度的改进，并在利用定位公式进行泄漏点定位时采用逐步迭代逼近算法，提高了系统的定位精度；
利用小波变换的多分辨率分解和重构技术，对采集到的管道发生泄漏时产生的负压波信号进行了降噪处理；并利用小波变换技术对信号奇异点的快速、准确的识别能力，对经上述消噪后的负压波信号进行了奇异点捕捉，精确捕捉的结果提高了定位公式中上下游站点采集到的负压波奇异点的时间差△t的精度，从而进一步提高了定位的精度；
考虑到小波变换是一种基波可变的信号分析工具，也就是说，不同的小波基波对信号分析的结果将有很大差别，这势必影响最终的处理结果，因此，我们又对比分析了本课题中小波基及尺度的选择问题，依据规则性系数相似性原理对母小波的选择进行了探索性研究；
最后，还提出了基于SCADA的泄漏检测与定位系统的软件开发构想及系统的嵌入方案。
As one of the five main transport industries, long pipeline delivery system and its safe operation are of great significant importance due to the serious consequences that may result from different reasons. Leaks from the pipelines can not only lead to a great deal of product losses and environment pollution, but also cause many people hurt, even to die. In recently, with some artificial destruction such as stealing natural gas and so on, leakage cases became more and more seriously. So it's very important and very necessary to find the leak and locate it on time in order to reduce the losses and keep the system operate normally.
With the "West Development" project carried on. the natural gas pipeline in our country will get into another high tide. Plenty of pipelines will be launched into building and running. In order to reduce the harm caused by leakage to the least and to get to higher leak sensitivity and location accuracy, it's very necessary to research the leak detection and location technique.
Seeing that the math model of natural gas pipeline is limited to the fluid character, terrain status, pipeline itself character as well as some complicated restriction, it's difficult to establish the accurate mathematic model. On the other hand, the SCADA systems have been installed in the new laid natural gas pipeline. Lots of local date can be easily obtained. At the same time, combining with the actual condition of delivery pipeline of our country and some prevalent leak detection and location methods, this paper raised a method, which is based on wavelet analysis and negative- pressure-wave to detect and locate the leakage. Jobs in the paper are as follows:
As affected by gas density, pressure, specific heat and pipe material, the velocity of negative-pressure-wave improved depending on the energy conservation. At one time, taking the affection of velocity of gas flow, the location formula was modified in some degree.
When locating the leaks with location formula, the iterative approach algorithm was adopted as a new method to increase the location accuracy level of system.
Multi-scale decomposition and reconstruction of wavelet transform is utilized to filter the noise of negative-pressure-wave, which was brought by leakage. With the capacity to detect the signal singularity fast and accurately, wavelet transform was used to catch the break point of pressure signal. The work improved the accuracy of At in location formula, accordingly improved the location accuracy.
Wavelet is a suitable tool for signal analysis. The distinct character is that the mother-wavelet is of optional. That means choosing different mother-wavelet will get different analysis result. So we studied how to choose mother-wavelet and its scale for signal singularity detection. The method is to choose a wavelet by regularities of the wavelet and the signal.
Finally, a plan, which bases on SCADA system for inspection and detection of natural gas leakage, is proposed in the article.
鉴于天然气长输管道的模型受到流体特性、地势状况及管道自身的特性等诸多复杂因素的制约，难于建立其精确的数学模型，新铺设的输气管网均已配有SCADA系统，对现场各类数据有了丰富的储备等因素，同时结合我国管道输送的实际情况，并借鉴目前国际上各种泄漏检测与定位的先进方法，针对天然气长输管线泄漏故障的特点，本文提出了基于小波分析和负压波相结合的检测泄漏与定位技术。其中，主要进行了以下几方面的工作：
针对泄漏产生的负压波在管道中的传播速度受媒介的密度、压力、比热及材质等因素的影响，而非声波在空气中传播的速度，本文利用能量守恒原理，对其进行了修正；考虑到管内气体流速对压力波速的影响，对原始的负压波定位公式进行了一定程度的改进，并在利用定位公式进行泄漏点定位时采用逐步迭代逼近算法，提高了系统的定位精度；
利用小波变换的多分辨率分解和重构技术，对采集到的管道发生泄漏时产生的负压波信号进行了降噪处理；并利用小波变换技术对信号奇异点的快速、准确的识别能力，对经上述消噪后的负压波信号进行了奇异点捕捉，精确捕捉的结果提高了定位公式中上下游站点采集到的负压波奇异点的时间差△t的精度，从而进一步提高了定位的精度；
考虑到小波变换是一种基波可变的信号分析工具，也就是说，不同的小波基波对信号分析的结果将有很大差别，这势必影响最终的处理结果，因此，我们又对比分析了本课题中小波基及尺度的选择问题，依据规则性系数相似性原理对母小波的选择进行了探索性研究；
最后，还提出了基于SCADA的泄漏检测与定位系统的软件开发构想及系统的嵌入方案。
As one of the five main transport industries, long pipeline delivery system and its safe operation are of great significant importance due to the serious consequences that may result from different reasons. Leaks from the pipelines can not only lead to a great deal of product losses and environment pollution, but also cause many people hurt, even to die. In recently, with some artificial destruction such as stealing natural gas and so on, leakage cases became more and more seriously. So it's very important and very necessary to find the leak and locate it on time in order to reduce the losses and keep the system operate normally.
With the "West Development" project carried on. the natural gas pipeline in our country will get into another high tide. Plenty of pipelines will be launched into building and running. In order to reduce the harm caused by leakage to the least and to get to higher leak sensitivity and location accuracy, it's very necessary to research the leak detection and location technique.
Seeing that the math model of natural gas pipeline is limited to the fluid character, terrain status, pipeline itself character as well as some complicated restriction, it's difficult to establish the accurate mathematic model. On the other hand, the SCADA systems have been installed in the new laid natural gas pipeline. Lots of local date can be easily obtained. At the same time, combining with the actual condition of delivery pipeline of our country and some prevalent leak detection and location methods, this paper raised a method, which is based on wavelet analysis and negative- pressure-wave to detect and locate the leakage. Jobs in the paper are as follows:
As affected by gas density, pressure, specific heat and pipe material, the velocity of negative-pressure-wave improved depending on the energy conservation. At one time, taking the affection of velocity of gas flow, the location formula was modified in some degree.
When locating the leaks with location formula, the iterative approach algorithm was adopted as a new method to increase the location accuracy level of system.
Multi-scale decomposition and reconstruction of wavelet transform is utilized to filter the noise of negative-pressure-wave, which was brought by leakage. With the capacity to detect the signal singularity fast and accurately, wavelet transform was used to catch the break point of pressure signal. The work improved the accuracy of At in location formula, accordingly improved the location accuracy.
Wavelet is a suitable tool for signal analysis. The distinct character is that the mother-wavelet is of optional. That means choosing different mother-wavelet will get different analysis result. So we studied how to choose mother-wavelet and its scale for signal singularity detection. The method is to choose a wavelet by regularities of the wavelet and the signal.
Finally, a plan, which bases on SCADA system for inspection and detection of natural gas leakage, is proposed in the article.
引文
[1] 曹辉玲，天然气管道的检漏工程，真空，1999．10．N0．5，pp．29
[2] 臧铁军等，我国油气管道运输的发展概况，油气储运，1997 16(7)，pp．1-3
[3] Pan Jia Hua,The development of China's Long-Distance Pipeline Industry, Pipe and Pipeline International,September-October, 1984,pp.7-9.
[4] 潘家华，管道输送的经济分析，油气管道技术动态，1980，pp1-12
[5] 杨祖佩，推进管道技术发展的有关问题的研究，管道技术与设备，1994(6)
[6] 周国雄，管道工业综述，国外油气储运，pp60-63，1990(1)
[7] 王霖，我国管道运输已成为五大运输行业之一，石油消息 1995年6月7日
[8] 潘家华，油气管道的风险分析(待续) 油气储运 1995 14(3)，pp．11-15
[9] Furness. R.A, Development in Pipeline Instrumentation, Measurement and Control, 1987, 20(1), pp. 7-15
[10] 近八年来俄罗斯油气管道重大事故大事记 石油消息 1995年6月14日
[11] 王迈编译 国外油气管道事故分析 世界石油工业 1996，3(12)，pp．45-52
[12] Alan Reeve, SCADA for Pipeline Monitoring and Control, C&I, November 1987, pp. 55-57
[13] R. Isermann, Process Fault Detection Based on Modeling and Estimation Methods-A survey, Automatica, Vol. 20, No. 4, pp387-404,1984
[14] Toshio Fukuda, Leak Detection and its Location in a pipeline System, Proceedings of IMEKO Symposium on Flow Measurement and Control in Industry, pp. 193-198, Nov, 1979
[15] A. Benkherouf, Leak Detection and Location in Gas Pipelines, IEEE Proceedings, Vol.135, No. 2, pp. 142-148, March, 1988
[16] Xue Jun Zhang, Statistical Leak Detection in Gas and Liquid pipelines, Pipes& Pipelines International, pp26-29, July-August 1993
[17] Witness, Sarah L. Gassman and M. Hanif Chaudhry, Leak Detection in Pipes by Frequency Response Method, Journal of hydraulic engineering, 127(2), pp134-147, February 2001
[18] Marco Ferrante, Bruno Brunone, Pipe System Diagnosis and Leak Detection by Unsteady-state Tests2 Wavelet analysis", Advances in Water Resources 2003 26, pp. 107-116
[19] 方崇智等，基于现代估计理论的长输管道自动监测，油气储运，1988(6)，pp．7-12
[20] 王桂增等，基于Kullback信息测度的长输管线的泄漏检测，信息与控制，1989(1)，pp．14-17
[21] 董东，王桂增，Kalman滤波器在长输管道泄漏诊断中的应用，自动化学报，1990，16(4)，pp．303-309
[22] 靳世久，强环境下噪声地下管道泄漏检测，天津大学学报，1994(5)
[23] 唐秀家，彦大春，基于神经网络的管道泄漏检测仪器及方法，北京大学学报，1997，33(3)，pp．319-326
[24] 陶洛文等，以辨识为基础的长输管线故障定位，清华大学学报，1986，26(2)，
pp．69-75
[25] 王海生等，基于小波分析的输油管道泄漏检测，信息与控制，Oct，2002，31(5)，pp．456—460
[26] X.J. Zhang, Designing a Cost-effective and Reliable Pipeline Leak Detection System, Pipes & Pipelines International,1997 January-February. pp.1-11
[27] 常景龙，李铁，输气管道泄漏检测技术的选择和优化，油气储运，2000，19(5)，pp．9—13
[28] N. C.Butter, BSC, Pipeline leak detection techniques, Pipes &Pipelines International, April, 1982
[29] 张布悦，王贵增等，输油管线泄漏检测与定位技术综述，上海海运学院学报，Sep．2001，22(3)，pp．13-16
[30] Donald A Abernathy, Airborne Detection of Pipe Line Leaks, Pipe Line Industry, 1987, 67(3),pp. 26
[31] 王福明，胡志新，相关分析在油气管道检漏中的应用，油气田地面工程，199817(5)，pp．13-20
[32] 黄文，毛汉领，包家福等，互相关分析在水管泄漏检测中的应用，广西民族学院学报，2001，7(4)，pp．??
[33] S M B, Beck, N J Willianmson, Pipeline system identification through cross-correlation analysis, ImechE Journal, 2002, Vo1216, Part E:J Process Mechanical Engineering, pp. 133-142
[34] H E Emara-Shabaik, Y A K hulief, I Hussaini, A non-linear multiple-model state estimation scheme for pipeline leak detection and isolation, ImechE Journal 2002, Vol 206 Part I:J System and Control Engineering, pp. 497-512
[35] 唐秀家，不等温长输管道泄漏定位理论，北大学报(自然科学版)，1997，33(5)，pp．574—579
[36] 王立宁，李健，靳世久，热输油管道瞬态压力波法泄漏点定位研究，石油学报，2000，21(4)，pp．93—96
[37] 李炜，陈希平，毛海杰等，天然气管道泄漏点的定位检测方法研究，甘肃工业大学学报，2003，29(4)，pp．84—87
[38] 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，电子测量与仪器学报，1998，12(1)，pp．59-64
[39] Ye H, Wang G Z, Fang C Z, Application of wavelet transform to leak detection and location transport pipelines, Engineering Simulation, 1996,13
[40] 李炜，潘纬，基于小波变换的输气管道泄漏检测与定位方法研究，计算机工程与应用，2002，9
[41] 周小勇，叶银忠，小波分析技术在故障诊断中的应用，上海海运学院学报，2001，22(3)，pp．116—119
[42] Dr Jun Zhang, Michael Twomey, Statistical pipeline leak detection techniques for all operating conditions, 26th Environmental Symposium & Exhibition, California, March, 2000
[43] 唐秀家，管道系统泄漏检测神经网络与模式识别方法，核科学与工程，1998，18(3)
[44] Salvatore Belsito, Paolo Lombardi, Paolo Andreussi, Sanjoy Banerjee,Leak detection in liquefied gas pipelines by artificial neural networks, AIChE Journal Dec 1998, 44 (12), pp. 2675-2687
[45] 姚光镇，输气管道设计与管理[M]，山东东营：石油大学出版社，1991
[46] 李玉柱，苑明顺．流体力学[M]，北京：高等教育出版社，1992。
[47] 冯耀荣．未来十年我国油气管道用钢管发展中的几个课题，焊管，2002，25(4)，pp．1-2
[48] 杨福生，小波变换的工程分析与应用[M]，北京：科学出版社，2000
[49] 胡昌华等，基于MATLAB的系统分析与设计，西安：西安电子科技大学出版社，1999
[50] Simon Hyaline, The wavelet Transform: A method for time-frequency localization. Advances in Spectrum Analysis and Array Processing, 1991
[51] 张兆宁，廖一原等，缓变信号奇异性的小波变换检测及其应用，系统工程理论与实践 2000，10(10)，pp．84-88
[52] 周小勇，叶银忠，故障信号检测的小波基选择方法，控制工程，2003，10(4)，pp．308-311
[53] 江汉秀，李琳等，长输管道自动化技术[M]，西安：西北工业大学出版社，2000
[54] Liou C. P., "Pipeline leak detection based on mass balance", Pipeline Infrastructure II, Proceedings of the International Conference, ASCE, 1993
[55] Marco Ferrante, Bruno Brunone, "Pipe system diagnosis and leak detection by unsteady-state testsl Harmonic analysis", Advances in Water Resources 26(2003):95-105
[56] G.Z. Wang, C.Z. Fang, K.F. Wang, State Estimation and Leak Detection in Pipeline IECON 1991
[57] C. Verde, Leak Location in Pipelines by Minimal Order Nonlinear Observer, Proceeding of the American Control Conference Arlington, VA, June 25-27 2001
[58] Kenya Fukushima, Reiko Maeshima, Gas pipeline leak detection system using online simulation method, Computers and Chemical Engineering 24(2000):453-456
[59] 周东华，叶银忠，现代故障诊断与容错控制[M]，北京；清华大学出版社，2000
[2] 臧铁军等，我国油气管道运输的发展概况，油气储运，1997 16(7)，pp．1-3
[3] Pan Jia Hua,The development of China's Long-Distance Pipeline Industry, Pipe and Pipeline International,September-October, 1984,pp.7-9.
[4] 潘家华，管道输送的经济分析，油气管道技术动态，1980，pp1-12
[5] 杨祖佩，推进管道技术发展的有关问题的研究，管道技术与设备，1994(6)
[6] 周国雄，管道工业综述，国外油气储运，pp60-63，1990(1)
[7] 王霖，我国管道运输已成为五大运输行业之一，石油消息 1995年6月7日
[8] 潘家华，油气管道的风险分析(待续) 油气储运 1995 14(3)，pp．11-15
[9] Furness. R.A, Development in Pipeline Instrumentation, Measurement and Control, 1987, 20(1), pp. 7-15
[10] 近八年来俄罗斯油气管道重大事故大事记 石油消息 1995年6月14日
[11] 王迈编译 国外油气管道事故分析 世界石油工业 1996，3(12)，pp．45-52
[12] Alan Reeve, SCADA for Pipeline Monitoring and Control, C&I, November 1987, pp. 55-57
[13] R. Isermann, Process Fault Detection Based on Modeling and Estimation Methods-A survey, Automatica, Vol. 20, No. 4, pp387-404,1984
[14] Toshio Fukuda, Leak Detection and its Location in a pipeline System, Proceedings of IMEKO Symposium on Flow Measurement and Control in Industry, pp. 193-198, Nov, 1979
[15] A. Benkherouf, Leak Detection and Location in Gas Pipelines, IEEE Proceedings, Vol.135, No. 2, pp. 142-148, March, 1988
[16] Xue Jun Zhang, Statistical Leak Detection in Gas and Liquid pipelines, Pipes& Pipelines International, pp26-29, July-August 1993
[17] Witness, Sarah L. Gassman and M. Hanif Chaudhry, Leak Detection in Pipes by Frequency Response Method, Journal of hydraulic engineering, 127(2), pp134-147, February 2001
[18] Marco Ferrante, Bruno Brunone, Pipe System Diagnosis and Leak Detection by Unsteady-state Tests2 Wavelet analysis", Advances in Water Resources 2003 26, pp. 107-116
[19] 方崇智等，基于现代估计理论的长输管道自动监测，油气储运，1988(6)，pp．7-12
[20] 王桂增等，基于Kullback信息测度的长输管线的泄漏检测，信息与控制，1989(1)，pp．14-17
[21] 董东，王桂增，Kalman滤波器在长输管道泄漏诊断中的应用，自动化学报，1990，16(4)，pp．303-309
[22] 靳世久，强环境下噪声地下管道泄漏检测，天津大学学报，1994(5)
[23] 唐秀家，彦大春，基于神经网络的管道泄漏检测仪器及方法，北京大学学报，1997，33(3)，pp．319-326
[24] 陶洛文等，以辨识为基础的长输管线故障定位，清华大学学报，1986，26(2)，
pp．69-75
[25] 王海生等，基于小波分析的输油管道泄漏检测，信息与控制，Oct，2002，31(5)，pp．456—460
[26] X.J. Zhang, Designing a Cost-effective and Reliable Pipeline Leak Detection System, Pipes & Pipelines International,1997 January-February. pp.1-11
[27] 常景龙，李铁，输气管道泄漏检测技术的选择和优化，油气储运，2000，19(5)，pp．9—13
[28] N. C.Butter, BSC, Pipeline leak detection techniques, Pipes &Pipelines International, April, 1982
[29] 张布悦，王贵增等，输油管线泄漏检测与定位技术综述，上海海运学院学报，Sep．2001，22(3)，pp．13-16
[30] Donald A Abernathy, Airborne Detection of Pipe Line Leaks, Pipe Line Industry, 1987, 67(3),pp. 26
[31] 王福明，胡志新，相关分析在油气管道检漏中的应用，油气田地面工程，199817(5)，pp．13-20
[32] 黄文，毛汉领，包家福等，互相关分析在水管泄漏检测中的应用，广西民族学院学报，2001，7(4)，pp．??
[33] S M B, Beck, N J Willianmson, Pipeline system identification through cross-correlation analysis, ImechE Journal, 2002, Vo1216, Part E:J Process Mechanical Engineering, pp. 133-142
[34] H E Emara-Shabaik, Y A K hulief, I Hussaini, A non-linear multiple-model state estimation scheme for pipeline leak detection and isolation, ImechE Journal 2002, Vol 206 Part I:J System and Control Engineering, pp. 497-512
[35] 唐秀家，不等温长输管道泄漏定位理论，北大学报(自然科学版)，1997，33(5)，pp．574—579
[36] 王立宁，李健，靳世久，热输油管道瞬态压力波法泄漏点定位研究，石油学报，2000，21(4)，pp．93—96
[37] 李炜，陈希平，毛海杰等，天然气管道泄漏点的定位检测方法研究，甘肃工业大学学报，2003，29(4)，pp．84—87
[38] 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，电子测量与仪器学报，1998，12(1)，pp．59-64
[39] Ye H, Wang G Z, Fang C Z, Application of wavelet transform to leak detection and location transport pipelines, Engineering Simulation, 1996,13
[40] 李炜，潘纬，基于小波变换的输气管道泄漏检测与定位方法研究，计算机工程与应用，2002，9
[41] 周小勇，叶银忠，小波分析技术在故障诊断中的应用，上海海运学院学报，2001，22(3)，pp．116—119
[42] Dr Jun Zhang, Michael Twomey, Statistical pipeline leak detection techniques for all operating conditions, 26th Environmental Symposium & Exhibition, California, March, 2000
[43] 唐秀家，管道系统泄漏检测神经网络与模式识别方法，核科学与工程，1998，18(3)
[44] Salvatore Belsito, Paolo Lombardi, Paolo Andreussi, Sanjoy Banerjee,Leak detection in liquefied gas pipelines by artificial neural networks, AIChE Journal Dec 1998, 44 (12), pp. 2675-2687
[45] 姚光镇，输气管道设计与管理[M]，山东东营：石油大学出版社，1991
[46] 李玉柱，苑明顺．流体力学[M]，北京：高等教育出版社，1992。
[47] 冯耀荣．未来十年我国油气管道用钢管发展中的几个课题，焊管，2002，25(4)，pp．1-2
[48] 杨福生，小波变换的工程分析与应用[M]，北京：科学出版社，2000
[49] 胡昌华等，基于MATLAB的系统分析与设计，西安：西安电子科技大学出版社，1999
[50] Simon Hyaline, The wavelet Transform: A method for time-frequency localization. Advances in Spectrum Analysis and Array Processing, 1991
[51] 张兆宁，廖一原等，缓变信号奇异性的小波变换检测及其应用，系统工程理论与实践 2000，10(10)，pp．84-88
[52] 周小勇，叶银忠，故障信号检测的小波基选择方法，控制工程，2003，10(4)，pp．308-311
[53] 江汉秀，李琳等，长输管道自动化技术[M]，西安：西北工业大学出版社，2000
[54] Liou C. P., "Pipeline leak detection based on mass balance", Pipeline Infrastructure II, Proceedings of the International Conference, ASCE, 1993
[55] Marco Ferrante, Bruno Brunone, "Pipe system diagnosis and leak detection by unsteady-state testsl Harmonic analysis", Advances in Water Resources 26(2003):95-105
[56] G.Z. Wang, C.Z. Fang, K.F. Wang, State Estimation and Leak Detection in Pipeline IECON 1991
[57] C. Verde, Leak Location in Pipelines by Minimal Order Nonlinear Observer, Proceeding of the American Control Conference Arlington, VA, June 25-27 2001
[58] Kenya Fukushima, Reiko Maeshima, Gas pipeline leak detection system using online simulation method, Computers and Chemical Engineering 24(2000):453-456
[59] 周东华，叶银忠，现代故障诊断与容错控制[M]，北京；清华大学出版社，2000