Demand modeling for water networks calibration and leak localization
The success in the application of any model-based methodology (e.g. design, control, supervision) highly depends on the availability of a well calibrated model. There is no best or unique solution for the calibration problem as the methodologies are developed depending on which parameters have to be...
| Autor: | |
|---|---|
| Tipo de recurso: | tesis doctoral |
| Fecha de publicación: | 2016 |
| 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:2117/96319 |
| Acceso en línea: | https://hdl.handle.net/2117/96319 https://dx.doi.org/10.5821/dissertation-2117-96319 |
| Access Level: | acceso abierto |
| Palabra clave: | Water distribution networks Calibration Leakage Sampling design Singular value decomposition Xarxes de distribució d'aigua Distribució de sensors Descomposició en valors singulars Calibració Aplicacions reals Fuites Detectors de fuites Aigua -- Distribució -- Calibratge Àrees temàtiques de la UPC::Informàtica |
| Sumario: | The success in the application of any model-based methodology (e.g. design, control, supervision) highly depends on the availability of a well calibrated model. There is no best or unique solution for the calibration problem as the methodologies are developed depending on which parameters have to be calibrated and the final use of the model. The main objective in this thesis is to develop an adaptive water distribution network model which both calibrates its demands online and discerns between faults and system evolution. The calibration is focused on demands due to their daily variability and continuous evolution. The singular value decomposition is a powerful tool for solving the optimization problem. Additionally, the deep understanding of this tool allows to redefine the demand model. A novel demand model is proposed, where each individual demand is defined as a combination of demand components. These demand components are calibrated demand multipliers that represent the behavior of nodes in a determined geographical zone. The membership of each nodal demand to every demand component is produced naturally through the analysis of the singular value decomposition of the sensitivity matrix. The same analysis is also used to define the location of sensors for the calibration. The calibration in water distribution networks needs to be performed online due to the continuous evolution of demands. During the calibration process, background leakages or bursts can be unintentionally incorporated to the demand model and treated as a system evolution (change in demands). To solve that, a leak detection and localization approach to be coupled with the calibration methodology that identifies geographically distributed parameters is proposed. The approach consists in comparing the calibrated parameters with their historical values to assess if changes in these parameters are caused by a system evolution or by the effect of leakage. The geographical distribution allows to associate an unexpected behavior of the calibrated parameters (e.g. abrupt changes, trends, etc.) to a specific zone in the network. The set of methods proposed are exemplified through an academic dummy network to help the reader completely understand their fundamentals. Furthermore, three real water distribution networks situated in Barcelona and Castelldefels are used to evaluate the performance of the whole method with real systems and real data. The good results obtained show the potential of the developed method and the viability of the real-time calibration and leak detection and localization processes. |
|---|