College of Civil Engineering, Hefei University of Technology, Hefei, China.
Centre for Water Systems, Faculty of Environment, Science and Economy, University of Exeter, Exeter, UK.
Water Res. 2024 Nov 15;266:122354. doi: 10.1016/j.watres.2024.122354. Epub 2024 Aug 28.
Many researchers have addressed the challenge of optimal pressure sensor placement for different purposes, such as leakage detection, model calibration, state estimation, etc. However, pressure data often need to serve multiple purposes, and a method to optimize sensor locations with versatility for various objectives is still lacking. In this paper, a graph-based optimal sensor placement (GOSP) framework is proposed, which aims to provide a robust and all-purpose approach to identify critical points for pressure monitoring. By analysing the spatial variation frequencies of WDN pressures, the relationship between measurements and the global variation of original pressures is established. On this basis, the D-optimality criterion is adopted to formulate the objective of GOSP, which aims to maximize the information on the spatial distribution of pressures that can be obtained from measurements. The new-proposed objective ensures that the sensor locations are compatible with various application scenarios. The proposed method was applied to a real-life distribution network, and was compared with other optimal sensor placement methods oriented towards burst detection and pipe roughness calibration. Based on comparative studies in different scenarios including unknown pressure estimation, burst detection, and model calibration, the effectiveness and robustness of the proposed method have been proved.
许多研究人员已经针对不同目的（如泄漏检测、模型校准、状态估计等）解决了最佳压力传感器放置的挑战。然而，压力数据通常需要服务于多种目的，并且仍然缺乏一种具有多功能性的传感器位置优化方法。本文提出了一种基于图的最优传感器放置（GOSP）框架，旨在为识别压力监测的关键点提供一种稳健且通用的方法。通过分析 WDN 压力的空间变化频率，建立了测量值与原始压力全局变化之间的关系。在此基础上，采用 D-最优性准则来制定 GOSP 的目标，即最大化从测量中获得的有关压力空间分布的信息量。新提出的目标确保了传感器位置与各种应用场景兼容。该方法应用于实际的配水网络，并与面向突发检测和管粗糙度校准的其他最优传感器放置方法进行了比较。基于在未知压力估计、突发检测和模型校准等不同场景中的比较研究，证明了所提出方法的有效性和鲁棒性。
