Synthetic water distribution networks of Peru for leak localization

Dataset of 1147 synthetic water distribution networks laid out on the OpenStreetMap streets and Copernicus GLO-30 elevations of 428 cities, towns and rural settlements of Peru, with 255518 simulated leak scenarios and the results of four baseline leak localization methods.

Each network includes its EPANET 2.2 model, one-week demand patterns with a 15 min step and sensor configurations for densities of 1, 5, 10 and 20 % of the service nodes. Each scenario stores the nominal pressure, the pressure difference caused by the leak and the leak parameters (pipe, position, type, equivalent diameter, start, growth and flow). The networks are split by place into training (788), validation (177) and test (182) sets. The test scenarios cover two days with a 15 min step; the training and validation scenarios keep the last 4 h of the pressure difference.

Files: networks.zip, scenarios_train.zip, scenarios_validation.zip, scenarios_test.zip and the tables folder (network catalogue, verification checks, scenario summary and benchmark results). README.txt describes every file and key.

Street data: © OpenStreetMap contributors, distributed under the Open Database License (ODbL).
Elevation data produced using Copernicus WorldDEM-30 © DLR e.V. 2010-2014 and © Airbus Defence and Space GmbH 2014-2018 provided under COPERNICUS by the European Union and ESA; all rights reserved. The organisations in charge of the Copernicus programme by law or by delegation do not incur any liability for any use of the Copernicus WorldDEM-30.
Population: WorldPop 2020 (CC BY 4.0).