China
Africa
Explore chapters and articles related to this topic
Alerting System for Gas Leakage in Pipelines
Published in K Hemachandran, Shubham Tayal, Preetha Mary George, Parveen Singla, Utku Kose, Bayesian Reasoning and Gaussian Processes for Machine Learning Applications, 2022
Nilesh Deotale, Pragya Chandra, Prathamesh Dherange, Pratiksha Repaswal, Saibaba V More
In this paper, the support vector machine technique is applied to detect leakage and predict the location in the water distribution and pipeline network of CSIR-CEERI, Pilani. In order to improve the accuracy and pinpoint the leakages, both the pressure values at junctions and flow values at pipelines are extracted from the EAPNET tool and used as a feature for analyzing through SVM to identify leakages rather than only the pressure value. EPANET is a simulation tool for water hydraulics and pipeline networks [8]. EPANET simulates the water hydraulic behavior of the water distribution and pipeline network that can be run over a desired time period and generates hydraulic parameters like pressure and flow at all points for that period of time. SVM techniques are supervised learning algorithms that use the learning datasets to train the model and further predict the future values using the trained model. SVM tries to obtain an optimal separation hyperplane that separates the different classes of learning data vectors to either sides of the hyperplane. SVM tries to maximize the margin distance between the support vectors.
Direct computation of Variable Speed Pumps for water distribution system analysis
Published in Bogumil Ulanicki, Kalanithy Vairavamoorthy, David Butler, Peter L.M. Bounds, Fayyaz Ali Memon, Water Management Challenges in Global Change, 2020
E. Todini, M.E. Tryby, Z.Y. Wu, T.M. Walski
EPANET is a model to run static and pseudodynamic (or quasi-steady) simulations i i.e. a successive of steady regimes) of the hydraulic and quality water pipe systems. The model gives values of discharge in each pipe, pressure in each node, the water level in each reservoir and the concentration of chemical species through the net during the simulation period, with a certain discretization. The model can be used as an aid tool for analyses of management alternative strategies as different scenarios for pumps operation and reservoirs filling or emptying. EPANET integrates all the main infrastructures that constitute the supplying systems, namely, gravity and pump systems, valves le.g.. relief, pressure reducing, regulating, control and isolation valves), reservoirs (of fixed or variable level), for which it is possible to establish operating conditions. The model calculates the balance conditions, for a set of equations, by using the method of the gradient.
Effect of leakage location on pressure deficiency index of water distribution network
Published in ISH Journal of Hydraulic Engineering, 2023
Nidhi C Pandya, Reena Popawala, S.M. Yadav
Similar to the laboratory setup, the leakages were evenly distributed throughout the network. As per the first scenario, leakage was applied at node 2, which was the location nearest to the source, whereas, in the fourth scenario, the leakage was applied at node 20, located at the tail end of the network. In scenarios 2 and 3, leakage was applied at nodes 14 and 19, which were located at the middle of the network. Similar to the laboratory WDN setup, leakages at Kosad WDN were analyzed for the maximum demand period. The software EPANET 2.2 was used for this analysis. EPANET 2.2 is an updated and expanded open source water distribution system modeling tool to meet the vastly important needs of water utilities and the water community. The software is able to analyze the network with a pressure-dependent flow approach. The variation in the pressure with respect to varying leakages during the extended period simulation was not considered as it is not within the scope of this study.
Stress-testing water distribution networks for cyber-physical attacks on water quality
Published in Urban Water Journal, 2022
Dionysios Nikolopoulos, Christos Makropoulos
RISKNOUGHT leverages the recently released EPANET 2.2 (Rossman et al. 2020) through the usage of the WNTR Python package (Klise et al. 2017) for representing in detail any WDN. The new version of EPANET facilitates natively pressure driven analysis (PDA) equations. These produce realistic pressure deficient conditions which may result in service unavailability in a WDN, in contrast with the normal setting of demand driven analysis (DDA) equations of previous EPANET versions (Ciaponi and Creaco 2018). This is of paramount importance in disaster modelling and when assessing the effect of prolonged or severe cyber-physical attacks. The water quality solver of EPANET 2.2. is also compatible with the PDA solver, allowing the handling of a single-species water contaminant fate and transportation analysis along with the hydraulics of the network. This used to not be a seamless process in the past), requiring extensions such as EPANET-PDX (Seyoum and Tanyimboh 2014, 2016) and EPANET-PMX (Seyoum and Tanyimboh 2017).
Design and development of a web-based EPANET model catalogue and execution environment
Published in Annals of GIS, 2021
Tylor Bayer, Daniel P. Ames, Theodore G. Cleveland
Developed by the United States Environmental Protection Agency (EPA), EPANET is a powerful open-source software suite used throughout public and private industries to model water distribution networks. In the past 5 years alone, there have been over 7,000 references to EPANET in peer review journals. Many of these research efforts lend to optimizing and extending EPANET’s capabilities, making it applicable to a wider range of real-world modelling scenarios. Some highly cited research articles related to EPANET include: Uniformly Distributed Demand EPANET Extension(Menapace et al. 2018).Noniterative Application of EPANET for Pressure Dependent Modelling Of Water Distribution Systems (Sayyed, Gupta, and Tanyimboh 2015).OOPNET: An object-oriented EPANET in Python (Steffelbauer and Fuchs-Hanusch 2015).Quest for a New Solver for EPANET 2 (Burger et al. 2016).