At present, water waste has become a global concern. On the other hand, the amount of sweet water on the earth is fixed and limited but the demand for water is increasing. This, more than ever before, makes it nec essary to modify the consumption pattern. One of the most important consumption management activities is to decrease the uncounted water. Water leakage not only results in loss of good - quality water resources, but also pollutes the drinking water and in it s worst form brings about serious damages to people and building around the point of leakage. In this paper, a model is presented for determining the amount and location of leakage in water supply networks. In this model which uses a neural network improve d by the bat optimization algorithm, the amount and location of leakage in the network is determined by the minimum number of pressure - measuring. The proposed model is applied on the Poulakis network when several simultaneous leakages have occurred, and t he accuracy of the model is verified by the results.

Keywords