Multi Objective Approach For Leakage Reduction In Water Distribution Systems
Price
Free (open access)
Transaction
Volume
103
Pages
10
Published
2007
Size
414 kb
Paper DOI
10.2495/WRM070581
Copyright
WIT Press
Author(s)
R. Magini, I. Pallavicini & D. Verde
Abstract
This paper faces the problem of reducing leakages in water distribution systems (WDS) through dynamic heads control with pressure reduction valves (PRV). To achieve this aim a multi-objective optimization approach is followed for the location of control valves and for their setting during different working conditions. In particular, the objectives to be achieved are: 1) to minimize the leakages over all of the network, 2) to minimize the investment costs for the control devices. The objectives must be satisfied considering the variability of the nodal water demand. For a more realistic simulation of the WDS performance the implementation of a fully pressure-dependent leakage specification is required. The multi-objective optimization problem is tackled using a MOGA (Multi-Objective Genetic Algorithm) with a Pareto based approach. The hydraulic modelling of the network is performed with the software EPANET2. The nodal demands are considered to be uncertain input parameters and the most robust solution is also found. The procedure is here applied to the WDS of a small town not far from Rome afflicted by heavy losses. A reduction of about 50% of initial water losses is achieved. Keywords: water distribution systems, leakage reduction, pressure control, multi objective optimization, genetic algorithms. 1 Introduction Nowadays the problem of water losses control in WDS is more and more pressing owing to the growth of population, the increasing of water consumptions and at the same time because of the deterioration and the reduction of available drinkable water resources as a consequence of pollution and climate changes. Leakages in urban water networks can be a very high percentage of the
Keywords
water distribution systems, leakage reduction, pressure control, multi objective optimization, genetic algorithms.