Optical Multimode Fiber-Based Pipe Leakage Sensor Using Speckle Pattern Analysis

Jonathan Philosof, Yevgeny Beiderman, Sergey Agdarov, Yafim Beiderman, Zeev Zalevsky

Research output: Contribution to journalArticlepeer-review

Abstract

Water is an invaluable resource quickly becoming scarce in many parts of the world. Therefore, the importance of efficiency in water supply and distribution has greatly increased. Some of the main tools for limiting losses in supply and distribution networks are leakage sensors that enable real-time monitoring. With fiber optics recently becoming a commodity, along with the sound advances in computing power and its miniaturization, multipurpose sensors relying on these technologies have gradually become common. In this study, we explore the development and testing of a multimode optic-fiber-based pipe monitoring and leakage detector based on statistical and machine learning analyses of speckle patterns captured from the fiber's outlet by a defocused camera. The sensor was placed inside or over a PVC pipe with covered and exposed core configurations, while 2 to 8 mm diameter pipe leaks were simulated under varied water flow and pressure. We found an overall leak size determination accuracy of 75.8% for a 400 µm covered fiber and of 68.3% for a 400 µm exposed fiber and demonstrated that our sensor detected pipe bursts, outside interventions, and shocks. This result was consistent for the sensors fixed inside and outside the pipe with both covered and exposed fibers.

Original languageEnglish
JournalSensors
Volume23
Issue number20
DOIs
StatePublished - 22 Oct 2023

Keywords

  • leak detection
  • machine learning (ML)
  • optic fiber sensor
  • speckle pattern

All Science Journal Classification (ASJC) codes

  • Analytical Chemistry
  • Information Systems
  • Instrumentation
  • Atomic and Molecular Physics, and Optics
  • Electrical and Electronic Engineering
  • Biochemistry

Fingerprint

Dive into the research topics of 'Optical Multimode Fiber-Based Pipe Leakage Sensor Using Speckle Pattern Analysis'. Together they form a unique fingerprint.

Cite this