To evaluate non revenue water (NRW) and losses in water distribution networks a methodology is developed by applying "annual water balance" and "minimum night flow" analyses. In this approach the main NRW components such as leakage from reported and un-reported bursts and background leakage, with real or estimated data, enabling assessment of indices of leakage performance are evaluated. Also, a novel procedure is introduced in this paper that can determine the nodal and pipe leakage by using a hydraulic simulation model. Recognising the pressure dependency of leakage the total consumption is divided into two parts, one pressure dependent and the other independent of local pressure, and the hydraulic behaviour of the network is analyzed. A computer code is developed to evaluate all components of water losses based on the proposed methodology. For better representation of the results and management of the system, the outputs are exported to a GIS model. Using the capabilities of this GIS model, the network map and attribute data are linked and factors affecting network leakage are identified. In addition, the effects of pressure reduction are investigated. The model is illustrated by a real case study. The results show that the suggested model has overcome the shortcomings of the existing
In this paper two models are presented based on Data-Driven Modeling (DDM) techniques (Artificial Neural Network and neuro-fuzzy systems) for more comprehensive and more accurate prediction of the pipe failure rate and an improved assessment of the reliability of pipes.Furthermore, a multivariate regression approach has been developed to enable comparison with the DDM-based methods. Unlike the existing simple regression models for prediction of pipe failure rates in which only few factors of diameter, age and length of pipes are considered, in this paper other parameters such as pressure and pipe depth, are also included. Furthermore, an investigation is carried out on most commonly used mechanical reliability relationships and the results of incorporation of the proposed pipe failure models in the reliability index are compared.The proposed models are applied to a real case study involving a large water distribution network in Iran and the results of model predictions are compared with measured pipe failure data.Compared with the results of neuro-fuzzy and multivariate regression models, the outcomes of the artificial neural network model are more realistic and accurate in the prediction of pipe failure rates and evaluation of mechanical reliability in water distribution networks.
A technique for leakage reduction is pressure management, which considers the direct relationship between leakage and pressure. To control the hydraulic pressure in a water distribution system, water levels in the storage tanks should be maintained as much as the variations in the water demand allows. The problem is bounded by minimum and maximum allowable pressure at the demand nodes. In this study, a Genetic Algorithm (GA) based optimization model is used to develop the optimal hourly water level variations in a storage tank in different seasons in order to minimize the leakage level. Resiliency and failure indices of the system have been considered as constraints in the optimization model to achieve the minimum required performance. In the proposed model, the results of a water distribution simulation model are used to train an Artificial Neural Network (ANN) model. Outputs of the ANN model as a hydraulic pressure function is then linked to a GA based optimization model to simulate hydraulic pressure and leakage at each node of the water distribution network based on the water level in the storage tank, water consumption and elevation of each node. The proposed model is applied for pressure management of a major pressure zone with an integrated storage facility in the northwest part of Tehran Metropolitan area. The results show that network 438 S. Nazif et al.leakage can be reduced more than 30% during a year when tank water level is optimized by the proposed model.
scite is a Brooklyn-based organization that helps researchers better discover and understand research articles through Smart Citations–citations that display the context of the citation and describe whether the article provides supporting or contrasting evidence. scite is used by students and researchers from around the world and is funded in part by the National Science Foundation and the National Institute on Drug Abuse of the National Institutes of Health.