.png){kind=icon}
.png){kind=icon}
Return
Anomaly detection using continuous wavelet transforms and local active information storage scores: An application to water distribution networks
.png){kind=icon}
OA
AI
E
G
DOI：10.1016/j.envsoft.2026.107074.png)
.png){kind=icon}
.png){kind=icon}
Share.png){kind=icon}
Save
Abstract
En 中文
• Tailored CWT pipeline removes nonlinear, non-polynomial trends to reveal subtle incipient leaks. • Novel use of Local AIS (LAIS) identifies anomalies via localized predictability. • Achieved 94% accuracy and a 0% false alarm rate on the Hanoi benchmark. • Detected minute 9 l/s leaks that state-of-the-art methods failed to identify.
Keywords:
Time series anomaly detection
Nonlinear and non-polynomial time series
Water distribution Network's (WDN) leak detection
Local active information storage (LAIS)
Continuous wavelet transform (CWT)
AI Summary
Key information extracted from the uploaded paper, including a brief overview, abstract, background, key highlights, visual analysis, and future outlook.
Journal
E
IF:
4.6
Papers:
191
Citations:
0
Organization
No organization information available
Cited Papers