Water distribution network model calibration and continuous maintenance: Terrassa, a real application

Abstract

The WDN model calibration has been and it still is a research area of interest [1]. This process has been well studied and the different steps required were already established. The availability of more and more data encourages the idea of applying these methodologies since the more the state of the system is known the better the estimations will be. Nevertheless, the application of all the academic proposals and experiences in real networks involve some major handicaps. First of all, the data validation and reconstruction before using them for the model calibration and for the model use afterwards. There are some automatic tools for calibration but they focus on the last step of the process when the parameters have to be tuned. Some previous steps that require an ad hoc solution. These issues represent a gap between academia and real practice. This paper presents a successful experience carried out with the collaboration of the municipal water company of Terrassa (TAIGUA) with a research group located in the same city with a long experience in water research. The objective of the project is to improve the existing models, establish their potentials and propose strategies for their final adjustment. The results are different depending on the availability of data. The district metered areas (DMA) presented have different characteristics and are representative of interesting situations for other practitioners. The models that result will allow a better management of the network including energy optimization, leak detection and localization and quality supervision.

This work has been partially funded by TAIGUA through the collaboration project with cs2ac (ref. UPC C-11660) and data have been kindly provided by TAIGUA and by the Spanish State Research Agency (AEI) and the European Regional Development Fund (ERFD) through the projects SCAV (ref. MINECO DPI2017-88403-R)

Peer Reviewed

Postprint (author's final draft)