Incremental upgrading sensor placement methodology: Application to the leak localization in water networks
This paper presents a new methodology for sensor reallocation in large scale-systems considering fault isolation purposes. From an initial set consisting of a limited number of sensors already installed in certain locations of the system, the proposed methodology produces a new sensor placement wher...
| Autores: | , , , , |
|---|---|
| Tipo de recurso: | artículo |
| Fecha de publicación: | 2022 |
| País: | España |
| Institución: | Universitat Politècnica de Catalunya (UPC) |
| Repositorio: | UPCommons. Portal del coneixement obert de la UPC |
| Idioma: | inglés |
| OAI Identifier: | oai:upcommons.upc.edu:2117/369780 |
| Acceso en línea: | https://hdl.handle.net/2117/369780 https://dx.doi.org/10.1016/j.compchemeng.2021.107642 |
| Access Level: | acceso abierto |
| Palabra clave: | Sensors Sensor placement Sensor reallocation Water distribution networks Leak localization Àrees temàtiques de la UPC::Informàtica::Automàtica i control |
| Sumario: | This paper presents a new methodology for sensor reallocation in large scale-systems considering fault isolation purposes. From an initial set consisting of a limited number of sensors already installed in certain locations of the system, the proposed methodology produces a new sensor placement where some of the sensors are strategically reallocated in different but available places of the system. The procedure is posed as an optimization problem where the performance index is specific of the fault isolation method to be used. The algorithm that solves the problem is an incremental upgrading approach based in the Sequential Forward Floating Search algorithm and it combines a forward phase (where sensors are added sequentially) with a backward phase (where sensors can be individually removed from the original sensor placement). The proposed methodology is illustrated by means of its application to the problem of leak localization in Water Distribution Networks (WDN), where the particular placement of the pressure sensors has a great impact in the ability to isolate the leaky node. Also, since only a limited number of pressure sensors can be installed in some nodes, space interpolation techniques must be used to estimate the pressure in the other network nodes. The performance of the localization process is measured in terms of pipe distance from the estimated leaky node candidate to the real leaky node. The proposed methodology is applied to two District Metered Areas of the WDN of a metropolitan area of Spain, in simulation and using real measurements and leak scenarios. |
|---|