Avaliação dos parâmetros de calibração do algoritmo genético na detecção de vazamentos em rede de distribuição de água utilizando o método transiente inverso

Abstract

Water is a fundamental natural resource for the maintenance of all forms of life, and its distribution must be pro- moted with quality and without waste. The idea of improving water distribution systems has arisen from some re- searchers and it allows for the simulation of real networks through computational models capable of studying their behavior in different situations. Inverse Transient Analysis (ITA) optimized by a Genetic Algorithm (GA) was applied to a synthetic distribution network to identify and quantify leaks. At the same time, the effects of variation of the GA operators (population size, number of generations, elitism rate and mutation rate) were evaluated as to the ac- curacy of the solutions found. The performance of the ITA-GA model was tested and its effectiveness in the identification and quantification of leaks was verified. However, its processing time can be a limiting element of its use.