In many countries water losses can be much larger than 50%, with great economic losses because of the energy required for pumping and for the primary treatments that very often are required. Methodologies have been developed in order to identify the areas where losses are most expected, limiting the excavations as far as possible. To this end, the authors developed a methodology which requires the installation of a number of instruments on the network, in order to measure pressures and discharges; then, the demands at the nodes are changed by means of optimization methods and the network is simulated with a computer program, in order to match the readings of the instruments; losses are higher where the demands have been most increased. In previous papers, the authors tested different evolutionary methods to identify the areas where losses are most expected, comparing results from theoretical networks that were, from all the other aspects, exactly identical. In the real world, unfortunately, differences from the simulated the real networks are not limited to the discharge demands, but other uncertainties are present. The most evident uncertainty is related to pipe roughness, which can be considered a parameter of the model; sometimes even the diameters are listed wrong. In the paper, the robustness of the proposed method is tested using networks where roughness parameters and diameters are different, changing the roughness parameter in order to determine the influence of the errors in the evaluation of the areas where losses are most expected.
[1]
John H. Holland,et al.
Outline for a Logical Theory of Adaptive Systems
,
1962,
JACM.
[2]
Riccardo Poli,et al.
Particle swarm optimization
,
1995,
Swarm Intelligence.
[3]
Thomas M. Walski,et al.
Analyzing Water Main Replacement Policies
,
1990
.
[4]
Godfrey A. Walters,et al.
Scheduling of Water Distribution System Rehabilitation Using Structured Messy Genetic Algorithms
,
1999,
Evolutionary Computation.
[5]
Daniele B. Laucelli,et al.
Development of rehabilitation plans for water mains replacement considering risk and cost-benefit assessment
,
2006
.
[6]
John H. Holland,et al.
Adaptation in Natural and Artificial Systems: An Introductory Analysis with Applications to Biology, Control, and Artificial Intelligence
,
1992
.
[7]
Stefano Mambretti,et al.
Genetic Algorithms For Leak Detection In Water Supply Networks
,
2012
.
[8]
Stefano Mambretti,et al.
Evolutionary computation techniques to assess losses in water supply networks
,
2013
.
[9]
Jeffrey Horn,et al.
Handbook of evolutionary computation
,
1997
.
[10]
Dragan Savic,et al.
Rehabilitation strategies for water distribution networks: a literature review with a UK perspective
,
2000
.
[11]
David E. Goldberg,et al.
Genetic Algorithms in Search Optimization and Machine Learning
,
1988
.