The skeletonization of a water supply network model and its impact on the quality of simulated results

Abstract

The application of numerical models of water supply networks has become a common practice not only in Poland. The models may be characterized by several types of geometrical structure: skeletonized, basic and detailed. Each of these types has an individual application in the process of supporting operational decisions. For example, to successfully optimize and schedule the operation of pump stations, skeletonized models are highly recommended. Typically, skeletonized models reflect the geometrical structure and diameters only of transmission mains and selected distribution pipes. Such representation of the geometrical structure of the network may cause problems during the calibration process – simulated results may differ significantly from the measured values of tested parameters. Taking that into account, the paper presents a different approach for the creation of a skeletonized model: the use of a replacement model created through the simplification of a basic model. The model, representing all mains and distribution pipes and additionally most important service connections, was skeletonized to the schematic form. The skeletonization process was presented on the basis of a selected actual water supply network. The correctness of the simulation results obtained with the use of the simplified model was compared with the measured values from the monitoring system (65 pressure monitoring points and 47 flow rate measurement points). The compliance of the simulation results with the measurement results was assessed with the use of AWWA guidelines. The conducted comparison showed a high agreement of the results of calculations and measurements.