Water is a central source of life, hence the control of fresh water quality delivered in supply networks is of great interest. The water quality is possibly threatened by intentional or accidental contaminations of industries, farms, or individual persons. The municipal water supply networks are usually of large scale, which makes it impossible to prevent containments by physical security measures. In our research we address the problem of identifying sources of containments from given measurements of a few sensor stations. A fast and reliable determination of the sources allows to start activities to prevent further containment.

We introduce a discrete-continuous model for the containment detection problem in water networks. The flow of water is dynamically modelled by the transport partial differential equation (PDE). We derive a linear mixed-integer model by discretizing the dynamics of the PDE and by approximations to the cost functional. We apply our methods to the municipal fresh water networks of
Memphis and Macao.

Partners

• TU Kaiserslautern

Related publications

• Armin Fügenschuh, Simone Göttlich, Michael Herty, Water Contamination Detection , Andreas Oberweis et al. (Eds.),eOrganisation: Service-, Prozess-, Market-Engineering, 8. Internationale Tagung Wirtschaftsinformatik, Universitätsverlag Karlsruhe, pp. 501 – 518, 2007.