Partial Optimization of Water Distribution System Accounting for Multiobjective System Safety

Abstract

The paper presents an application of an evolutionary multi-objective optimization method in the estimation a Pareto optimal set of solutions of pipe replacement schedule. An optimization task is a problem with total investment cost, system entropy, hydraulic reliability and water losses as the objective functions. The paper shows that it is possible to optimize a part of a water distribution system without a deterioration of the crucial parameters in the rest of it, which also results in reduction of a computational complexity. Next, the paper presents how to select the pipes that have to be replaced by application of specific coding of decision variables. The results, based on real life example shows that the method is able to identify the pay-off surface characteristic between the proposed objectives.