Smart meter-powered water use behavior modelling and management: data, algorithms, and privacy-enhancing technologies
Over the last three decades, the development of smart water meter trials and the rise of “digital water” have enabled the collection and analysis of water demand data at increasingly higher spatial and temporal resolutions. Leveraging smart water meter data to build knowledge on when and how water is used in different households up to the fixture level yields valuable insights for demand management and efficiency programs for both customers and water providers. Yet, many controversial aspects related to the uncertain technological, economic, and social benefits of high-resolution digital meters still limit the availability of household-level water usage data for research and practice. In this talk, we will first analyze the development of advanced metering infrastructure in the global context. Second, we will build on recent research efforts developed on data across three continents to show how water consumption data at different spatial and temporal scales, coupled with advanced data analytics, can be used to improve our understanding of water usage behaviors at different spatiotemporal scales and assist demand management programs, with demonstrative case studies in Europe, Australia, and the USA. Finally, we’ll quantitatively analyze how we can protect users’ privacy and avoid exposing sensitive information about their water activities, especially if an adversary gains access to such data, while guaranteeing sufficient data access for research and planning purposes.
Andrea Cominola is Assistant Professor of Smart Water Networks at the Einstein Center Digital Future and Technische Universität Berlin. In early 2017 he received a PhD in Information Technology at Politecnico di Milano, following his BSc cum laude in Environmental Engineering at Politecnico di Milano in 2010 and a double-degree MSc cum laude at Politecnico di Milano and Politecnico di Torino (Alta Scuola Politecnica) in 2013. He is leading a group with 5 PhD students and research assistants, and 2 graduate students, and he is involved in a number of national and international projects, raising more than 3.4 M€ since the start of the Smart Water Networks group in 2018. From February 2017 to March 2018 he has been Post- doctoral research fellow in the Natural Resources Management group of Professor Andrea Castelletti at the Department of Information, Electronics and Bioengineering (Politecnico di Milano) and from April 2018 to September 2018 Post-doctoral Research Assistant at the Chair of Fluid System Dynamics of the Technische Universität Berlin. In early 2016, he has been visiting PhD candidate at the Watershed Sciences Center and Center for Water-Energy Efficiency of UC Davis (CA - USA). He was formerly visiting researcher at Korea University (Seoul, Korea), and visiting student at PennState University (PA - USA) and Norwegian University of Science and Technology (Trondheim, Norway). Dr. Cominola’s research focuses on urban water and energy demand modelling and management, leakage and anomaly detection, behavioral modelling, data mining, and machine learning for water and coupled human-environment systems analysis. He is author of 23 scientific publications in peer-reviewed international journals, 2 book chapters, more than 75 publications/presentations in international conferences, and reviewer for several international journals, organizations, and institutions. In 2022, he co-authored a technical report on ”Climate change, infrastructure and mobility. Solutions and strategies for infrastructure investments in a context adaptation to climate change and mitigation of greenhouse gas emissions”, prepared for the Italian Ministry of Sustainable Infrastructures and Mobility.
Since May 2021 he is the elected Chair of the HS5 subdivision on Water policy, management and control at the European Geosciences Union (EGU). Since 2019 he serves as Associate Editor for the Journal of Water Resources Planning and Management (American Society of Civil Engineers) and, since 2023, for Water Resources Research (American Geophysical Union) and npj Clean Water