Study of a flow model for detection and location of leaks and obstructions in pipes - SYSCO Accéder directement au contenu
Communication Dans Un Congrès Année : 2012

Study of a flow model for detection and location of leaks and obstructions in pipes

Résumé

This paper presents the development of a flow model for the final purpose to detect and locate leaks and obstructions in pipes. This model is based on one-dimensional water hammer partial differential equations, discretized with an explicit finite-difference method, to represent the unsteady incompressible flow in pipes, for a finite number of positions. First a study on the selection of boundary conditions for the best structure to represent a pipe system is proposed. This study intends to improve previous works based on some 'simplified' finite-difference scheme and basically using pressure boundary conditions at both sides of the pipe (at the beginning and at the end). In particular all configurations of boundary conditions are reviewed, alternating between pressure and flow at each end. Then substitution of those boundary conditions by pump and flow restriction equations is considered, in order to obtain a more realistic model and consider the possibility to measure less variables. Some simulations are finally presented and compared with real data, illustrating how the pipe behavior is better represented with the new boundary conditions, pumps and restriction equations. Those behaviors are considered in particular in the presence of leaks, making the model of interest for possible use in leak detection and isolation.
Fichier non déposé

Dates et versions

hal-00741054 , version 1 (11-10-2012)

Identifiants

  • HAL Id : hal-00741054 , version 1

Citer

Marcos Guillén, Jean-François Dulhoste, Gildas Besancon, Rafael Santos. Study of a flow model for detection and location of leaks and obstructions in pipes. MOSIM 2012 - 9th International Conference of Modeling, Optimization and Simulation, Jun 2012, Bordeaux, France. pp.n/c. ⟨hal-00741054⟩
161 Consultations
0 Téléchargements

Partager

Gmail Facebook X LinkedIn More