Online leakage detection system based on EWMA-enhanced Tukey method for water distribution systems

Abstract Real-time leakage detection based on pressure and flow data has become increasingly essential for water distribution systems (WDSs). Recent data-driven approaches have largely focused burst characterised as sudden outflow or drops but did not mention the ability to detect gradual events that do change could cause more loss. This study proposes an online system exponential weighted moving average (EWMA)-enhanced Tukey method help monitor of WDSs. The proposed comprises three main parts: pre-processing, sub-system, parameter updating sub-system. is lightweight powerful statistical tools without complex model construction. effectiveness demonstrated datasets under various real-world scenarios, including leakages bursts. results showed EWMA-enhanced quickly while generating low false alarms. computationally effective able deal with non-stationary behaviours automatically.