Prediction model for the leakage rate in a water distribution system

Abstract

Leakages cause real losses in water distribution systems (WDSs) from transmission lines, storage tanks, networks, and service connections. In particular, the amount of leakage increases in aging networks due to pressure effects, resulting in severe water losses. In this study, various artificial neural network (ANN) models are considered for determining monthly leakage rates and the variables that affect leakage. The monthly data, which are standardized by Z-score for the years 2016–2019, are used in these models by selecting four independent variables that affect the leakage rate regarding district metered areas and pressure metered areas in WDSs. The pressure effects are taken into consideration directly as input. The model accuracy is determined by comparing the predicted and measured data. Furthermore, the leakage rates are estimated by directly modelling the actual data with ANNs. Consequently, it is found that the model results after data standardization are somewhat better than the original nonstandardized data model results when 30 neurons are used in a single hidden layer. The reason for the higher accuracy in the standardized case compared with previous modelling studies is that the pressure effect is taken into consideration. The suggested models improve the model accuracy, and hence, the methodology of this paper supports an improved pressure management system and leakage reduction.

Understanding district metered area level leakage using explainable machine learning

“…This is important to avoid 'leaking' information about the test set back into the training set. This method of normalization is similar to the 'Z-score' in [6,9] but replaces the mean and standard deviation with the median and interquartile range.…”

“…Other research has estimated non-revenue water ratios and leakage rates [6][7][8][9]. Many of these works use principal component analysis and artificial neural networks [6][7][8][9]. These methods, while more sophisticated, are less explainable than the methods used in this paper.…”

“…Other research has estimated non-revenue water ratios and leakage rates [6][7][8][9]. Many of these works use principal component analysis and artificial neural networks [6][7][8][9].…”