摘要
提出一种城市供水管网水压监测点多目标优化布局方法,通过在现有供水管网中合理布置水压监测节点,有效提高城市供水管网漏失事件识别效率。表征管网节点异常波动的压力残差向量是评估漏失事件的一个重要依据,不同漏失点造成的压力残差向量之间夹角值可以用来表征管网漏失节点的可识别性能。考虑模型输出不确定性造成的影响,引入漏失特征空间理论,漏失特征域能够刻画由模型输出不确定度带来的压力残差向量波动。当两个特征域发生重叠时,意味着两个漏失点的漏失特征十分相近,存在误识别区分漏失点的风险。同时对漏失特征域重叠程度跟漏失特征域重心夹角均值两个目标寻优,确定最优布局方案。本文以某城市供水管网系统为实例展开分析,验证了基于多目标的优化布局方法在科学合理的指导监测点选址工作以及提升管网漏失可识别性能方面的实用性。
This work presents a multi-objective optimal sensor placement approach to improve the leak detection in water distribution networks. The pressure residual vectors is an important basis for assessing the leakage event. The angle between the pressure residual vectors can be used to characterize the identifiable performance of leakage points. It is difficult to analyze the impact of the uncertainty of the model output by the method above. Therefore, the leakage signature space(LSS) theory is introduced, the model output uncertainty caused by the fluctuations can be covered by the leakage feature domain. When two domains overlap, means that the two leakage characteristics are very similar, there is a risk that the leakage point can’t be identified properly. The optimal sensor placement is determined by simultaneously optimizing the size of the overlapping signature domains and the mean value of the barycenters of the leakage signature domain. In this paper, an urban water distribution network is taken as an example to verify that the multi-objective optimization layout method can guide the selection of the monitoring site and effectively improve the identifiability of the monitoring system.
作者
周全
谢翔
侯迪波
黄平捷
张光新
张宏建
ZHOU Quan;XIE Xiang;HOU Di-bo;HUANG Ping-jie;ZHANG Guang-xin;ZHANG Hong-jian(State Key Laboratory of Industrial Control Technology,College of Control Science and Engineering,Zhejiang University,Hangzhou 310027,China)
出处
《控制工程》
CSCD
北大核心
2020年第7期1169-1174,共6页
Control Engineering of China
基金
国家自然科学基金资助项目(U1509208,61573313)
浙江省科技厅重大科技专项资助项目(2015C03014)。
关键词
供水管网
优化选址
水压监测点
多目标优化
漏失定位
Water distribution networks
optimal sensor placement
pressure monitoring points
multi-objective optimization
leakage detection
