供水管网水力状态模拟及爆管信号识别研究

发布时间：2018-04-03 23:43

本文选题：供水管网　切入点：爆管　出处：《昆明理工大学》2017年硕士论文

【摘要】：供水管网是城市重要的基础设施之一,近年来各国供水行业发展迅速。随着供水管网覆盖面积的不断扩大,管网爆管漏损现象也在不断增加。针对供水管网爆管的研究俨然已成为各个领域的研究热点。本文在分析了以往供水管网的爆管研究成果后,发现在供水管网爆管信号识别方面的研究还较为空白且目前仅有的一些方法的识别精度仍然不高,因此本文提出了基于CUSUM方法的供水管网爆管信号识别研究。首先,提出采用软件构建管网水力模型。近年来,供水管网水力模型在供水管网设计、运营以及调度等方面的应用越发广泛,因此,针对管网实际运行数据获取困难这一问题,本文通过对现有软件对比分析最终选用Water GEMS软件来构建实例管网的水力模型从而通过仿真获取所需数据。其次,结合国内外学者的研究成果和实例管网的具体用水情况分析了目前用户用水的规律,最后结合时间序列分析方法和模特卡罗模拟方法对用水的特性曲线进行了探究并在此基础上对供水管网的正常运行进行了仿真。然后,对目前学者采用的渗漏模型和渗漏面积计算公式进行了分析。根据小孔自由出流理论并考虑水流外渗时土体的阻碍作用,对本文采用的渗漏模型进行了推求。最后根据推求的结果进行了管网爆管仿真。最后,由于CUSUM方法具有能对较小异常做出明显反应、计算简单、反应灵敏、滤波等优点,本文创造性的将CUSUM方法用于爆管信号的识别中,以期提高识别精度。对CUSUM方法的应用研究具体包括了以下内容:CUSUM方法分析、数据稳定性处理、数据正态检验、正态转换以及实例应用。其中,在进行数据稳定性处理时创造性的提出提取单一时刻用水量组成时间序列,从而降低了用水正常波动的影响,在保证数据的稳定性的同时也从根本上提高了识别精度。最后,实例证明,利用CUSUM方法能够提高爆管信号识别精度并有效的避免误报。本文所提方法可以为供水管网的相关研究提供参考。
[Abstract]:Water supply network is one of the most important infrastructure in the city, and the water supply industry has developed rapidly in recent years.With the continuous expansion of water supply network coverage, pipe leakage phenomenon is also increasing.The research on pipe burst of water supply network has become a research hotspot in every field.After analyzing the previous research results of pipe burst in the water supply network, it is found that the research on the signal recognition of the pipe burst in the water supply network is still relatively blank and the recognition accuracy of only some methods is still not high.Therefore, this paper presents a research on the signal recognition of pipe burst in water supply network based on CUSUM method.Firstly, the software is used to construct the hydraulic model of pipe network.In recent years, the hydraulic model of water supply network is more and more widely used in the design, operation and dispatching of water supply network.By comparing and analyzing the existing software, Water GEMS software is used to build the hydraulic model of the example pipe network to obtain the needed data through simulation.Secondly, combined with the research results of domestic and foreign scholars and practical examples of the specific water situation of pipe networks, the current law of user water use is analyzed.Finally, the characteristic curve of water supply is explored with time series analysis method and model Carlo simulation method, and the normal operation of water supply network is simulated on this basis.Then, the seepage model and leakage area formula used by scholars are analyzed.According to the theory of small hole free discharge and considering the blocking effect of soil when the flow outside the seepage, the seepage model adopted in this paper is deduced.Finally, the pipe burst simulation is carried out according to the calculated results.Finally, due to the advantages of CUSUM method, such as simple calculation, sensitive response, filtering and so on, the CUSUM method is creatively used in the recognition of tube burst signals in order to improve the recognition accuracy.The research on the application of CUSUM method includes the following contents: analysis of the CUSUM method, data stability processing, data normality test, normal transformation and application of examples.Among them, in the process of data stability processing, it is creatively proposed to extract the time series of water consumption at a single time, thus reducing the influence of the normal fluctuation of water use, and at the same time improving the recognition accuracy fundamentally while ensuring the stability of the data.Finally, an example shows that the CUSUM method can improve the accuracy of tube burst signal recognition and effectively avoid false positives.The method proposed in this paper can be used as a reference for the research of water supply network.
【学位授予单位】：昆明理工大学
【学位级别】：硕士
【学位授予年份】：2017
【分类号】：TU991.33

【参考文献】