A fuzzy-stochastic modelling approach for urban water supply systems – ISS EWATUS project’s concepts.

---

Rafal Ulanczyk, Katarzyna Samborska, Wojciech Froelich, Ewa Magiera, Chrysi Laspidou, Jose Salmeron

Thursday 2 july 2015

9:30 - 9:45h at North America (level 0)

Themes: (T) Special session, (ST) FP7 ICT and water

Parallel session: 10I. Special session: FP7 ICT and water

---

According to the Organisation for Economic Co-operation and Development, water quality deterioration and extreme weather events will be the main factors that threaten the freshwater resources in Europe. In the near future, all European countries will be affected by water shortages or extreme weather events that cause droughts or floods. In these circumstances the elaboration of new water-saving procedures and policies are extremely important. The ISS EWATUS project integrates two areas of research: ICT (information and telecommunication technology) and public water management. The main goal is the development of an intelligent integrated support system for efficient water use and resources management. The project is carried out at two levels of different scales: household and urban. In both cases, the final product will be a decision support system (DSS) that will help to optimise water use. At urban scale the DSS has to be based on the reliable and calibrated model of the water distribution network. One of most common tools used for simulating water supply systems is the Epanet. It is widely applied in water distribution and water quality studies. Epanet is a deterministic tool so in order to incorporate the uncertainty of hydraulic properties that may affect the accurate estimation of flow and pressure conditions in pipes the additional stochastic module has to be employed e.g. by using Epanet Toolkit for programmers. Moreover, water demand being the crucial parameter of the water supply model will be forecasted using the fuzzy logic method. By means of such approach the uncertainties of flow parameters and water demand will be minimised. Accordingly, the Epanet model will be executed in two modes: near real time and forecasting mode.