基于时间序列分析的供水管网协同控制方法研究

发布时间：2018-01-28 02:13

本文关键词： 供水网络物联网时间序列分析 ARIMA模型预测控制分布式控制　出处：《江苏大学》2017年硕士论文　论文类型：学位论文

【摘要】：近年来,随着我国经济的高速发展和国民生活水平的不断提高,大量的高层建筑被建设,公众对生活用水的水质和水压都提出了更高的要求。无负压供水因为能够很好地减少二次供水的污染和有效提高水压,已经得到广泛的应用。二次供水设备的运维保障问题、节能问题及减少供水设备对城市供水主管网带来的影响问题,成为目前供水行业亟待解决的典型问题。精确的预测控制能够有效地降低二次供水设备对城市供水主管网带来的压力冲击,减少供水主管网的压力波动,提高供水主管网及其周边供水支管网的运行的稳定性。论文以物联网技术为基础,通过时间序列分析及预测控制等技术对远程供水设施进行协同控制。主要工作包括以下几个方面:首先,论文综合了供水基础设施特点和物联网技术,实现了对供水设施的远程数据采集及对供水设备的远程控制。设计了数据采集硬件模块,采用Modbus、以太网和GPRS的方式实现了采集器与PLC及采集器与服务器之间的数据通信,实现了采集器从PLC读取供水的实时数据、采集器将供水状态数据上传至服务器及服务器向采集器下发远程控制命令。并且通过采集器上的短信发送功能,确保了当供水管网有异常发生时,服务器能够及时通知远程采集器发送报警信息,提高供水管网的运维效率,有效降低运维成本。其次,论文对从供水设施所采集的数据进行了时间序列分析。根据供水管网的日用水数据时间序列具有明显的周期性与趋势性,提出了基于季节性ARIMA模型的供水数据预测模型。利用该模型对供水数据进行预测,所得预测结果与实际值的误差小,具有较为理想的预测精度。针对有爆管等异常发生时水压通常会产生较大变化的特征,提出通过计算不同采集点之间的水压的异常程度,来对异常进行判定与定位。然后,论文在讨论协同控制的方法和技术的基础上,叙述了对供水管网进行爆管检测的常见方案,提出将独立计量区域(DMA)及分布式模型预测控制技术(DMPC)用于供水管网的远程协同控制,并完成了模型设计和仿真。最后,介绍了协同控制系统的拓扑结构及整体架构,并对供水管网的预测控制方案的实现效果进行了分析。结果表明,本系统能够有效地对供水管网进行预测控制,在供水行业上具有良好的应用前景,并且该方案在其他的类似场景中具有可适应性。
[Abstract]:In recent years, with the rapid development of China's economy and the continuous improvement of national living standards, a large number of high-rise buildings have been built. The public put forward higher demands on the water quality and water pressure of domestic water. Non-negative pressure water supply can reduce the pollution of secondary water supply and raise water pressure effectively. The operation and maintenance of secondary water supply equipment, energy saving and reducing the impact of water supply equipment on urban water supply network have been widely used. Accurate predictive control can effectively reduce the pressure impact of secondary water supply equipment on urban water supply network and reduce the pressure fluctuation of water supply network. To improve the stability of the water supply network and its surrounding branch water supply network. The paper is based on the Internet of things technology. Collaborative control of remote water supply facilities is carried out through time series analysis and predictive control. The main work includes the following aspects: firstly, this paper synthesizes the characteristics of water supply infrastructure and Internet of things technology. The remote data acquisition of water supply facilities and remote control of water supply equipment are realized. The hardware module of data acquisition is designed and Modbus is adopted. Ethernet and GPRS realize the data communication between collector and PLC and between collector and server, and realize that the collector can read the real-time data of water supply from PLC. The collector uploads the water supply state data to the server and the server and sends the remote control command to the collector, and through the short message sending function on the collector, it ensures that when the water supply network has abnormal occurrence. The server can notify the remote collector in time to send alarm information, improve the efficiency of operation and maintenance of water supply network, and effectively reduce the cost of operation and maintenance. This paper analyzes the time series of the data collected from the water supply facilities. According to the time series of the daily water use data of the water supply network, it has obvious periodicity and trend. A prediction model of water supply data based on seasonal ARIMA model is proposed, which is used to predict the water supply data, and the error between the predicted results and the actual values is small. In view of the characteristics that the water pressure usually changes greatly when the tube burst occurs, it is proposed to calculate the abnormal degree of the water pressure between different sampling points. Then, on the basis of discussing the method and technology of cooperative control, the common scheme of pipe burst detection for water supply network is described. An independent measurement area (DMA) and distributed model predictive control technology (DMPC) are proposed for remote collaborative control of water supply network, and the model design and simulation are completed. This paper introduces the topological structure and the whole structure of the collaborative control system, and analyzes the effect of the predictive control scheme of the water supply network. The results show that the system can effectively predict and control the water supply network. It has a good application prospect in water supply industry, and the scheme has adaptability in other similar scenarios.
【学位授予单位】：江苏大学
【学位级别】：硕士
【学位授予年份】：2017
【分类号】：TP273;TU991.33

【参考文献】