Detection of Pipes Decreasing Residual Chlorine Via Wall Reaction Coefficient in Water Distribution Networks

Authors : Miraç Eryiğit

Pages : 86-94

View : 51 | Download : 27

Publication Date : 2024-11-25

Article Type : Research Paper

Abstract :This paper intended to build an optimization model utilizing the modified clonal selection algorithm (one of the famous heuristic optimization techniques) to detect pipes which reduces a residual chlorine in the water distribution networks (WDNs). MATLAB programming language was used to code the model linked with EPANET. The model performance was evaluated in a two-loop hypothetical WDN under steady-state flow conditions. In nodes of this hypothetical WDN, free chlorine concentrations were assumed to be measured since an objective function depends on model calibration. Pipes decreasing residual chlorine concentrations were determined by running the model which minimizes a total of concentration differences between estimated and measured free chlorine in each node. In order to find these pipes, pipe wall reaction rate coefficients were utilized. The model was run 10 times to obtain average reaction rate coefficient of each pipe in the WDN. After 10 runs, mean estimated and actual/real reaction rate coefficient values were almost equal (R2=0.99). The optimization model appeared to be viable for detecting pipes causing a residual chlorine loss in the WDN.
Keywords : Yapay bağışıklık sistemleri, model kalibrasyonu, boru cidar reaksiyon katsayısı, bakiye klor, su dağıtım şebekesi

ORIGINAL ARTICLE URL