Management of Water Loss Based of Bayesian Networks

Abstract

The losses that occur in water pipes installations can lead to the fact that some quantity of water does not reach the consumers, due to possible damages occurred in the distribution system. Bayesian networks and their associated methods are especially suited for capturing and dealing with uncertainty. A reason for this paper is that developing systems for solving problems, having the complexity present in monitoring the water loss, is still a major task. An effective way for saving water and money is represented by the localization and solving the damages that appear in a pipe-network. The structure of the models is achieved using probabilistic (chance) nodes and directed links (structure). The parameters (influences between variables) are quantified and stored in conditional probability tables.We propose an interactive Bayesian network and a decision theoretic system which intend to monitor the water loss, to likely predict the outcome and to select the appropriate decisions. We consider that the model proposed in this paper offers a good way of monitoring the water distribution system. This way, isolations and remedies for the damages are possible.
Copyright © 2014 Praise Worthy Prize - All rights reserved.