数据挖掘技术在DMA产销差系统中的应用

发布时间：2018-02-23 18:29

本文关键词： DMA 数据挖掘回归分析聚类分析统计分析产销差　出处：《福建师范大学》2016年硕士论文　论文类型：学位论文

【摘要】：水是生命之源,城市供水管网肩负着向城市的各个角落输送自来水的重大使命,是一个城市的命脉所在。然而,随着经济社会的高速发展,供水管网产销差远未达到预期,一方面与人们的节水意识较差有关,另一方面与供水管网存在漏损有关。同时,各种偷水、漏水现象的存在,不仅浪费了宝贵的水资源,也浪费了大量的人力物力财力资源。因此,降低管网产销差是关系国计民生的大事,是供水行业急需迫切解决的问题。供水管网分区管理技术(DMA)技术已成为供水行业公认有效的管网漏损控制手段,对控制管网漏损、降低供水产销差具有重大意义。随着DMA技术在国内外各大城市的广泛应用,企业和自来水公司可以采集到大量的用户用水数据,基于大量的真实用水数据,采用统计技术和数据挖掘技术进行处理和挖掘,从表面的用水数据挖掘背后的用水规律和行为,生成各类数据报表,推送给企业领导层,提供辅助决策支持。本文以福州市部分小区独立分区计量的用户用水数据为数据源,采用最小流量分析,回归分析、聚类分析等方法分析和挖掘各独立分区计量区域的总表和户表用水规律和行为,为解决产销差问题中的用水量预测、夜间最小流量的精确计算等问题提供了新的视角,并根据研究成果运用编程方面的知识建立了DMA系统。本文的主要工作内容及贡献如下：(1)研究现状和存在的问题的总结和分析介绍了课题的背景与研究意义,详述了DMA技术的国内外发展现状及存在问题,重点阐述了数据挖掘技术中的回归分析,聚类分析方法。(2)夜间最小流量的精确计算针对传统的最小夜间流量计算的误差较大问题,调研相关的文献研究,采用基于正态分布理论的夜间最小流量计算方法,较为准确的求取夜间最小流量,能较好的辅助判断分区计量小区管网的漏损情况。(3)用水量预测根据自来水公司对用水量预估的需要,分析历史用水数据,发现用水量总体上的趋势性变化,采用回归模型进行分析,建立预测模型,预测各小区的短期的用水量。(4)各区域用水行为的聚类分析与评估对各独立分区计量小区,从用水量,产销差,夜间最小流量等维度进行聚类分析,发现各小区用水的规律,并进行客观的用水评估。(5)DMA系统的设计与实现针对分区计量的特点,设计和部署了DMA产销差系统,包括系统的架构,总体功能设计,数据库设计,主要功能模块的实现。
[Abstract]:Water is the source of life, urban water supply network shoulders the major mission of supplying tap water to every corner of the city, and is the lifeblood of a city. However, with the rapid development of economy and society, the difference in production and sale of water supply network is far from expected. On the one hand, it has something to do with people's poor consciousness of saving water, on the other hand, it is related to the leakage of water supply networks. At the same time, the existence of various water theft and leakage phenomena not only wastes valuable water resources, It also wastes a lot of human, material and financial resources. Therefore, reducing the gap between production and sales in pipe networks is a major event related to the national economy and the people's livelihood. It is an urgent problem to be solved urgently in the water supply industry. DMA-technology has become an effective and effective method to control the leakage of the water supply network, and it can control the leakage of the pipe network. With the wide application of DMA technology in domestic and foreign cities, enterprises and water companies can collect a large number of user water data, based on a large number of real water data. Using statistical technology and data mining technology to deal with and mining, from the surface of water data mining behind the law and behavior of water use, generate all kinds of data reports, push to the enterprise leadership, In this paper, based on the user water data measured in part of the district of Fuzhou as data source, minimum flow analysis, regression analysis, Clustering analysis and other methods are used to analyze and excavate the rule and behavior of water consumption in the main table and household table of the metering area in each independent area, which provides a new angle of view for solving the problems of water consumption prediction and accurate calculation of minimum flow rate at night in the problem of producing and marketing difference. According to the research results, a DMA system is established by using the knowledge of programming. The main contents and contributions of this paper are as follows: 1) the current research situation and existing problems are summarized and analyzed, and the background and significance of the research are introduced. In this paper, the present situation and existing problems of DMA technology at home and abroad are described in detail, and the regression analysis and clustering analysis method in data mining technology are emphasized. The accurate calculation of minimum nocturnal flow is aimed at the large error of traditional minimum nocturnal flow calculation. Based on the theory of normal distribution, the nocturnal minimum flow is calculated accurately. According to the need of water supply company to estimate water consumption, the historical water use data are analyzed, and the trend of water consumption is found. The regression model is used to analyze and establish the prediction model, and the cluster analysis and evaluation of the water use behavior of each district are analyzed and evaluated. The water consumption, production and marketing difference are analyzed and evaluated. By cluster analysis of minimum flow at night and other dimensions, the rule of water use in each district is found, and the design and implementation of DMA system based on the characteristics of zonal measurement are carried out, including the structure of the system. General function design, database design, main function module realization.
【学位授予单位】：福建师范大学
【学位级别】：硕士
【学位授予年份】：2016
【分类号】：TU991.6;TP311.13

【相似文献】