Analysis of water use characteristics in public buildings using high-precision metering data: a case study in Beijing

ABSTRACT A graphic summary of the experimental procedure, including the data collection and analysis processes. The experiment used smart water meters to collect four types of data, which were then analyzed in both the frequency and time domains. Digital twins have become an important means for achieving intelligent and refined management of water use in urban public buildings. However, traditional water metering systems cannot provide the high-resolution data required for digital-twin applications and refined operation. To address this gap, we conducted a 500-day field study in a research institute in Beijing, collecting high-resolution water-use data across multiple scenarios using smart water meters and survey instruments. The results show that water-use patterns in public buildings are highly uncertain and complex, influenced by multiple factors at different levels. Most water anomalies were repairable malfunctions in squat toilets, typically occurring during daytime hours and leading to substantial leakage. Water consumption varied significantly by time and date, with pronounced reductions on statutory holidays, and weekend water use was markedly lower than weekday use. The weekday–weekend gap in women's water consumption exceeded that of men. Gender-related differences were primarily associated with squat-toilet use, and women were more likely to use sinks due to additional influencing factors.These findings provide essential data and characteristic knowledge for developing digital twins of water systems in public buildings.