Federated knowledge integration and machine learning in water distribution networks

Water distribution networks are geographically distributed systems, with greater heterogeneity in terms of control structures, management strategies, and varying geometry with continuous expansion and changes in demand along their life. Due to these characteristics, water distribution companies face the problem of data and knowledge integration related with control and optimal exploitation. A few European and International RTD projects are currently focused on the design and development of a next generation system to support the control, optimal operation and decision support of drinking water distribution networks. In the context of a two year RTD project - WATERNET*, an evolutionary knowledge capture for advanced supervision of water distribution network is being developed. The WATERNET system, assists the distributed control of water management network to minimize the costs of exploitation, guarantee the continuous supply of water with a better quality monitoring, save energy consumption and minimize natural resources waste. This system comprises of several subsystems: a distributed information management subsystem, a machine learning subsystem, an optimization subsystem, a water quality monitoring subsystem, a simulation subsystem, and a supervision system that integrates these subsystems in order to assist the decision making and optimal operation of the network. This paper first provides a high level global description of this on going research on water management system. Then, it concentrates on the current stage of development on two specific subsystems in the project; the distributed information management and the machine learning subsystems.