【作者】 胡琼；

【导师】 范世东；

【作者基本信息】 武汉理工大学 ， 载运工具运用工程， 2009， 博士

【摘要】 随着工业的不断发展,管路被广泛应用于船舶、大型重装设备及石油、天然气等输送系统。船舶管路包括燃油管路、滑油管路、冷却水管路、压载水管路等等,其中任一管路发生泄漏,都可能导致相关设备无法正常使用,甚至影响到主辅机的安全使用;大型重装设备中的管路通常为油管和水管,以电站锅炉为例,电站锅炉“四管”(水冷壁、过热器、再热器和省煤器)爆漏事故是影响火电机组安全、经济运行的重要因素;石油和天然气等工业管道的泄漏不仅造成严重的资源浪费和环境污染,甚至会威胁到周围居民的生命财产安全。因此,展开对在役管线泄漏检测技术的研究具有重要的学术意义和工程实用价值。本文建立了一套管线泄漏检测综合试验台,并在此基础上进行了稳态输送和非稳态输送下的管线泄漏检测试验研究,最终完成了对管线泄漏检测试验系统的误差分析。本文的主要工作和研究内容如下:(1)简单介绍了常用的各种管线泄漏检测技术的基本原理,选用负压波—流量联合法进行管线泄漏的检测与定位,并采用小波分析作为管道压力和流量信号的数据处理方法,其中,利用小波降噪对信号消噪,利用小波变换奇异点分析检测负压波信号和流量信号的突变点,从而实现泄漏的检测与定位。(2)搭建了一套综合的管线泄漏检测试验台,模拟液体输送管线正常运行以及发生泄漏时的状况,并以负压波—流量联合法为基本原理,采用虚拟仪器LabVIEW开发了一套管线泄漏检测试验系统。该系统不仅能实时监测管线的运行状况,及时检测泄漏的发生并进行泄漏点的定位,还能通过查询时间实现历史数据的回放。在该试验台上,能开展不同泄漏量、不同管输状态(稳态输送和非稳态输送)、不同管线布局(水平输送管道和垂直输送管道)等多种方案的泄漏检测试验研究。(3)在试验台的基础上,讨论了管流处于稳定输送状态时,影响泄漏检测定位精度的因素。分别对一段直管和一段带有弯头且管线布局发生变化的复杂管道展开了泄漏检测试验,试验方案采用正交试验设计法设计,通过对试验结果的方差分析,得出了如下结论:对于直管,管输量越小,泄漏定位的精度越高,而泄漏量对泄漏定位精度也略有影响;对于复杂管道,管输量和泄漏量的不同会影响到泄漏定位精度,同时管道中的弯头数量越多,泄漏定位精度越差。(4)在试验台的基础上,开展了非稳态输送下的管线泄漏检测试验研究。刚开启泵时,管输流体压力波动较大,此时管流处于非稳定输送状态,对四种不同工况下刚开启泵时管流的压力波动程度进行了分级,压力波动程度越小时泄漏定位的精度越好,通过对不同压力波动程度下的试验数据进行分析,得出了非稳定输送状态下进行泄漏检测时,泵开启后不同阶段的泄漏定位修正值,提高了泄漏定位精度。(5)分析了管线泄漏检测试验系统的误差来源。随机误差主要是由各种干扰噪声引起,包括共模干扰、串模干扰和信号通道干扰。系统误差包括传感器的温度误差、非线性误差和灵敏度误差,数据采集中的采样误差和A/D转换误差,以及计算机与检测软件中的计算机运算误差、显示误差和数据处理误差。最后给出了管线泄漏检测试验系统的误差构成公式,并分别讨论了修正或抑制随机误差、传感器误差、数据采集误差和计算机与检测软件误差的方法。更多还原

【Abstract】 With the development of industry, pipelines are widely used in ship, major equipment, oil and gas transport system. Ship pipe path includes fuel pipe, lube pipe, cooling water pipe and so on, and the leak of any pipe may cause related equipments unable to use, even influence the safe use of main and auxiliary engines; pipelines in major equipments are usually oil pipe and water pipe, taking utility boiler for example, the accident of "four-tube" crack in boiler has always been one of the major problems in power plant, which baffles the unit operation safely and economically; the leak of oil and gas pipeline causes not only large resource waste and huge environmental pollution, but also threatens people’s lives and property. Therefore, it has important significance in research and engineering application to study on leak detection technology for in-service pipeline.In this paper, a pipeline leak detection testing system had been developed, a series of pipeline leak detection testing research were carried out, and error analysis of pipeline leak detection testing system had been completed. Follows are the main works:(1) The fundamental principle of common pipeline leak detection methods are simply introduced, negative pressure wave method and flow method are combined to detect and locate pipeline leak, wavelet analysis is chose as signal processing method of pipeline pressure and flow, in which, wavelet denoising method is used to filter away noise from original pressure and flow signals and wavelet singularity analysis is used to detect catastrophe points of pipeline pressure and flow signals, so pipeline leak detection and location can be implemented.(2) A pipeline leak detection synthetical testing platform was built in the laboratory to simulate normal operation and leak condition of liquid delivery pipeline, and a set of pipeline leak detection testing system was developed by Virtual Instrument LabVIEW, which based on negative pressure wave and flow. This system can not only monitor operation condition of the pipeline real-time, detect and locate leak in time, but also replay history data by query time. Based on the testing platform, a variety of leak detection testing research can be carried out under different leakage rate, different pipeline transportation condition such as steady transportation condition and unsteady transportation condition and different pipeline layout such as horizontal transport pipeline and vertical transport pipeline.(3) On the basis of the testing platform, when pipe flow is under steady transportation condition, factors having influence on leak location precision were discussed. Leak detection tests were carried out for a section straight pipe and a section complicated pipeline which has elbows and layout of beam line changes. The test plan was developed by orthogonal test design method. Through the variance analysis of testing results, we can get conclusions as follows: for a section straight pipe, the small pipeline transport flux, the better the leak location precision is, and leakage rate has little influence on leak location precision; for a section complicated pipeline, pipeline transport flux and leakage rate have influence on leak location precision, and the more elbows in pipeline, the worse leak location precision is.(4) On the basis of the testing platform, pipeline leak detection tests when flow is under unsteady transportation condition were carried out. On pumps start-up, the liquid pressure oscillation is comparatively large; flow is under unsteady transportation condition at this time. Under four kinds of different conditions, pressure fluctuation is divided into three degrees, we can find that leak location precision is better when pressure fluctuation is small, and by analyzing testing data of different pressure fluctuation degree, a set of modified value were provided, leak location precision had been improved.(5) Error sources of pipeline leak detection experimental system had been analyzed. Random error was caused by various interference noises, including common mode interference, series mode interference and signal channel interference. Systematic error includes temperature error, nonlinear error and sensitive error of sensors, sampling error and analog-digital conversion error in data acquisition, computer calculation error, displaying error and data processing error of virtual instrument software. Error constitution formula of pipeline leak detection system had been summarized, and several methods were provided to amend and restrain errors.更多还原