Application of artificial neural networks to predict pipe failure in water supply networks

Abstract

Water leakage on the water supply network is still a problem to be solved by water supply companies. One of the main causes of this phenomenon is due to pipe failure. Prediction models of water pipe failure are also constantly being improved to accurately and quickly predict the position of pipe leakage. With the aim of developing a better prediction, this study proposed application of the neural network model to predict pipe failure on water supply network. The proposed model is verified by the water supply network in ward 17
(DMA17), Go Vap district, Ho Minh city. The quality of the proposed model was evaluated by using the AUC standard. The results shows that the proposed model is very good.