Burst Detection and Localization using Discrete Wavelet Transform and Cross-Correlation
Abstrak
Burst in water distribution systems causes great loss of this natural resource, interrupts the water supply, damages the streets, builds and increases the transmission of infectious diseases. In this paper we propose a new algorithm that allows the detection and automatic localization of burst in water distribution systems. As for detection, the novelty is to use the wavelet correlation criterion to compute the statistical decision and compare it with a detection threshold. The novelty in the localization is to use the statistical operator cross-correlation. The algorithm was implemented in Octave and was validated with 32 signals acquired in the laboratory in a 26.7 m long steel pipe. In 16 signals burst were triggered which were detected under a false positive probability of 2 %. No false positives were present on the 16 signals where only noise was present.
Topik & Kata Kunci
Penulis (4)
Eduardo Trutié-Carrero
Damian Valdés-Santiago
Ángela León-Mecías
Jorge Ramírez-Beltrán
Akses Cepat
- Tahun Terbit
- 2018
- Sumber Database
- DOAJ
- DOI
- 10.4995/riai.2017.8738
- Akses
- Open Access ✓