|M.Sc Student||Maskit Matan|
|Subject||Multi-Objective Optimization for Least Cost Operation|
and Leakage in Water Distribution Systems
|Department||Department of Civil and Environmental Engineering||Supervisor||Professor Avi Ostfeld|
|Full Thesis text|
This research is aimed at developing and solving a multi-objective water distribution systems problem incorporating pumps optimal scheduling and leakage minimization. The constructed model is multi-dimensional and nonlinear, combining pumps control, consumer demands, storage, and other water distribution network analysis features.
The two objectives (i.e., leakage and energy cost) are minimized simultaneously utilizing a multi-objective genetic algorithm (GA) optimization technique coupled with hydraulic simulations.
Results show that leakage minimization is competing against the tendency to pump at low night tariffs (especially when there are significant differences between demands during low and high tariffs). The model can thus enhance the decision making process of water distribution systems operators.
The developed model also deals with the calibration and computation of leakage, which might reach 55% of total supply. As such, leakage control has an important impact on system operation as it generally increases with pressure.
Several studies tried to pair the connection between background leakage, pressure, and pipe characteristics (e.g., age, material). In this work a method is presented for calibrating and computing the leakage with an iterative optimization technique. The calibration process is based on searching aggregated leakage parameters.
The model is demonstrated on two example applications with increasing complexity, through base runs and sensitivity analyses.