Towards sustainable water networks: automated fault detection and diagnosis
WATERNOMICS – Towards sustainable water networks: automated fault detection and diagnosis
Domenico Perfido, R2M Solution, Italy– is a registered professional engineer in Ita-ly in the field of civil engineering with specializations in hydraulic engineering, energy engineering and environmental engineering.
Chiara Zanotti, R2M Solution, Italy- holds a summa cum laude Master Degree in Environmental science form the Universita Milano Bicocca.
Niall chambers, NUI Galway, Ireland– is a graduate of NUI Galway (2014), in the new and innovative Energy Systems Engineering bachelor’s degree course, in which he specialised in Mechanical Engineering and achieved a high first class honours.
Louise Hannon, NUI Galway, Ireland– works as an employee at NUI Galway and as chartered engineer with 18 years’ experience in a wide variety of civil end environmen-tal engineering projects.
Marcus Keane, NUI Galway, Ireland– has extensive knowledge and experience in the development of integrated Building Information Models (BIM) that encapsulate the processes and data associated with holistic environmental & energy management in buildings and industrial processes.
Eoghan Clifford, NUI Galway, Ireland– is currently a lecturer in Civil Engineering, NUI Galway with 12 years’ experience in the areas of water, wastewater, waste treat-ment and sustainable transport in the academic, research and private spheres
Andrea Costa , R2M Solution, Italy- is a qualified chartered engineer and building energy rating (BER) assessor in Italy and a Certifiend Energy Manager (CEM) affiliated to the Association of Energy Engineers (AEE). He is expert in building simulation with experience on an array of building energy simulation software and ISO 50001 certifica-tion tools.
The novelty of this project is to apply to water systems field all the best practices and innovation already applied to the energy field. FDD has been applied recently in energy and HVAC systems but not yet to water systems and water network simulation environ-ment, this is done for the first time within this project.
This paper present one of the results of the WATERNOMICS project, the key problem addressed in WATERNOMICS is the lack of water information, manage-ment and decision support tools that pre-sent meaningful and personalized infor-mation about usage, price, and availabil-ity of water in an intuitive and interactive way to end users. On average water net-works in EU have leakages and ineffi-ciencies that results in 20-30% water losses, new technologies are needed to solve this issue and make EU more sus-tainable. The paper will give an overview of the WATERNOMICS project and pre-sent the of FDD approach for water net-works developed within the project. In particular, the FDD system developed is based on the hydraulic modeling of the water network, the real time values of flow and pressure obtained from installa-tion of innovative ICT and commercial smart meters and the application of the ADWICE algorithm adapted for the drinking water network. The FDD system developed is useful when more than one parameter need to be considered at the same time to determine if an anomaly or fault is in place in a complex water net-work. The system is designed on pur-pose to cope with a larger features set.
The new FDD system will be implement-ed in an Italian demo site, the Linate Airport Water network in Milan, where a large water distribution network is in place and where, due the many varia-bles coming into play, it could be very difficult to detect anomalies with a low false alarm rate. The anomaly detection module will be based on an existing al-gorithm, called ADWICE . This class of algorithms is based on modelling the system selecting the best set of parame-ters that characterize the operational conditions (flow rate, pressure, energy consumption for pumps system, ground water level for the wells, etc.) assuming normal operation, i.e. absence of prob-lems (leaks, faults, etc.). This model will be used as a comparison baseline with the operational values observed by the water sensors installed in the network in real time. Whenever the system under observation is not found to be operating in the modelled normal region and the deviation between the normality and the current situation exceeds a certain threshold, an alarm is raised.
 M. Raciti, “Anomaly Detection and its Adap-tation: Studies on Cyber-Physical Sys-tems,” Linköping University, 2013.
 M. Raciti, J. Cucurull, and S. Nadjm-Tehrani, “Anomaly Detection in Water Management Systems,” in Advances in Critical Infrastructure Protection: Infor-mation Infrastructure Models, Analysis, and Defense, 2012, pp. 98–119.
Ireland Pilot – Waternolics